Methods and apparatus to determine probabilistic media viewing metrics
Summary by NHIP
Probabilistic viewing metric apparatus
The apparatus calculates media viewing metrics using panelist probabilities derived from viewing data. It computes conditional probabilities based on device presence during a first time period and approximates viewership shares via a conditional distribution.
Claim Score by NHIP
Abstract
Methods and apparatus to determine probabilistic media viewing metrics are disclosed herein. An example apparatus for determining a viewing metric for media to be viewed by a plurality of panelists includes a probability identifier to identify a probability for respective ones of the panelists with respect to the panelists viewing the media. The probability identifier is to identify the probability based on viewing data for the respective ones of the panelists. The example apparatus includes a calculator to calculate the viewing metric for the media based on the probabilities for the respective ones of the panelists and a sampling weight assigned to the respective ones of the panelists.
Term
10.2 yearsleft in the term
Expires 20 December 2036.
- Priority and filed
- Granted
- Today
- Expires
19 claims: 3 independent, 16 dependent
- 1An apparatus for determining a viewing metric for media to be viewed by a plurality of panelists, the apparatus comprising:a probability identifier to identify a probability for respective ones of the panelists with respect to the panelists viewing the media, the probability identifier to identify the probability based on viewing data for the respective ones of the panelists, the viewing data including incomplete viewing data for one or more of the panelists relative to the media;and a calculator to: calculate a conditional probability for the respective ones of the panelists based on the probability for the respective ones of the panelists, the conditional probability based on a condition of the respective ones of the panelists viewing respective media presentative devices during a first time period associated with a presentation of the media;calculate a share weight for the respective ones of the panelists based on a sampling weight assigned to the respective ones of the panelists and a probability of the respective ones of the panelists not viewing the respective media presentation devices during the first time period;and approximate a share indicative of viewership of the media using a conditional distribution based on the conditional probability and the share weight calculated for the respective ones of the panelists.
- 9A method for determining a viewing metric for media to be viewed by a plurality of panelists, the method comprising:identifying, by executing an instruction with a processor, a probability for respective ones of the panelists with respect to the panelists viewing the media, the identifying based on viewing data for the respective ones of the panelists, the viewing data including incomplete viewing data for one or more of the panelists relative to the media;calculating, by executing an instruction with the processor, a conditional probability for the respective ones of the panelists based on the probability for the respective ones of the panelists, the conditional probability based on a condition of the respective ones of the panelists viewing respective media presentative devices during a first time period associated with a presentation of the media;calculating, by executing an instruction with the processor, a share weight for the respective ones of the panelists based on a sampling weight assigned to the respective ones of the panelists and a probability of the respective ones of the panelists not viewing the respective media presentation devices during the first time period;and approximating, by executing an instruction with the processor, a share indicative of viewership of the media using a conditional distribution based on the conditional probability and the share weight calculated for the respective ones of the panelists.
- 15Broadest claimClaim Score 52, average(NHIP)A non-transitory computer-readable medium comprising instructions that, when executed, cause a processor to at least:identify a probability for respective ones of panelists with respect to the panelists viewing a media, the processor to identify the probability based on viewing data for the respective ones of the panelists, the viewing data including incomplete viewing data for one or more of the panelists relative to the media;calculate a conditional probability for the respective ones of the panelists based on the probability for the respective ones of the panelists, the conditional probability based on a condition of the respective ones of the panelists viewing respective media presentative devices during a first time period associated with a presentation of the media;calculate a share weight for the respective ones of the panelists, based on a sampling weight assigned to the respective ones of the panelists and a probability of the respective ones of the panelists not viewing the respective media presentation devices during the first time period;and approximate a share indicative of viewership of the media using a conditional distribution based on the conditional probability and the share weight calculated for the respective ones of the panelists.
Independent claims3
110 paragraphs in 4 sections, as filed
FIELD OF THE DISCLOSURE
0001This disclosure relates generally to media viewing metrics such as ratings and shares and, more particularly, to methods and apparatus to determine probabilistic media viewing metrics.
BACKGROUND
0002Audience viewership of, for example, a television program, may be analyzed to determine ratings and/or shares for the program. Audience viewing behavior data collected from, for example, a viewing panel, may introduce uncertainties into the analysis of the ratings and/or shares. For example there may be uncertainties as to whether a panelist is watching television and, if so, what television channel or program the panelist is watching.
BRIEF DESCRIPTION OF THE DRAWINGS
<figref idref="DRAWINGS">FIG. 1</figref> illustrates an example environment in which a system constructed in accordance with the teachings disclosed herein operates.
<figref idref="DRAWINGS">FIG. 2</figref> is a block diagram of an example implementation of a portion of the system of <figref idref="DRAWINGS">FIG. 1</figref>.
<figref idref="DRAWINGS">FIG. 3</figref> is a flowchart representative of example machine readable instructions that may be executed to implement the example system of <figref idref="DRAWINGS">FIGS. 1-2</figref>.
<figref idref="DRAWINGS">FIG. 4</figref> illustrates an example processor platform that may execute the example instructions of <figref idref="DRAWINGS">FIG. 3</figref> to implement the example system of <figref idref="DRAWINGS">FIGS. 1-2</figref>.
0007The figures are not to scale. Wherever possible, the same reference numbers will be used throughout the drawing(s) and accompanying written description to refer to the same or like parts.
DETAILED DESCRIPTION
0008Audience viewing data can be collected from a plurality of individuals or households watching, for example, television, to determine ratings and/or shares for one or more television programs. Television ratings represent a number of people (or households) with a television tuned to a particular channel or program divided by a total number of people (or households) that have a television. Thus, ratings consider a potential viewing population, or the total number of people or households that have a television. Television shares represent a percentage of people (or households) watching a particular channel or program out of a viewing population that includes the people (or households) that are watching television at a given time. Thus, determining shares includes considering a population who is watching television at a given time.
0009When analyzing audience viewing behavior to calculate ratings and/or shares, there may be uncertainties with respect to whether a panelist (e.g., a person in a household selected to participate in ratings research performed by, for example, The Nielsen Company (US), LLC) is watching television and, if so, what television channel and/or program he or she is watching. Uncertainties with respect to identifying audience viewing behavior can arise from, for example, co-viewing of a television program by members of the same household or a malfunction of a television panel meter collecting viewing activity data from the panelist's television. Thus, in some examples, viewing metrics such as ratings and/or shares are determined using data including uncertainties or probabilities with respect to panelist viewing behavior.
0010For example, for a first panelist, there may be a 50% probability that the first panelist is not watching television or a 50% that the first panelist is watching a first program. As another example, data may be collected from a second panelist indicating that the second panelist is watching television, but there may not be data as to which of a first program, a second program, or a third program the second panelist is watching. Known methods for addressing probabilities or uncertainties with respect to the viewing behavior of, for example, the first panelist and the second panelist include randomly assigning each panelist as viewing a particular television program using a Monte Carlo simulation or a variation thereof. For example, the first panelist who is either not watching television or is watching the first program may be randomly assigned as watching the first program. The second panelist who is watching one of the first program, the second program, or the third program may be randomly assigned to the second program. Thus, in some known methods, each of the first panelist and the second panelist are assigned as watching a particular program or as not watching television (e.g., using “0's” and “1's”), thereby removing uncertainties from the panelist data.
0011In some known methods, ratings and/or shares can be calculated based on the randomly assigned probability data (e.g., the 0's and 1's) for the first panelist, the second panelist, and/or other panelists. However, in some known methods, a Monte Carlo simulation is only performed once. As a result, ratings information does not account for the fact that the panelists could be watching other programs probabilistically. For example, if the Monte Carlo simulation is performed multiple times, the second panelist could be randomly assigned as watching the first program or the third program instead of the second program. Thus, ratings and/or shares calculated based on random assignment of panelist viewing activity may not accurately reflect a range of possible probabilistic scenarios, as the results are limited by the different scenarios that are generated.
0012Accuracy of such known methods could be increased if, for example, the Monte Carlo simulation is performed multiple times (e.g., thousands of times) to identify a range of possible scenarios or outcomes with respect to probabilities that a panelist is watching television, what program a panelist is watching, etc. and if the ratings calculated from the different probabilistic scenarios are averaged. However, such known methods are time-consuming and can require significant processing resources to repeat the simulation thousands of times in an effort to capture a wide range of possible probabilistic scenarios or outcomes. Even if the simulation is run multiple times, the results are still limited by the fact that ideally the simulation would be run an infinite number of times.
0013Examples disclosed herein provide for a determination of viewing metrics such as ratings and/or shares that accounts for substantially all possible viewing scenarios that could happen and a probability of a viewing scenario happening. For example, ratings computed using examples disclosed herein consider that the second panelist could be watching the first program, the second program, or the third program as well as the respective probabilities that the second panelist is watching the one of first, second, or third programs. Examples disclosed herein compute ratings and/or shares for one or more television programs using one or more algorithms that consider the probabilities that a panelist may or may not be watching television, may or may not be watching a certain program, etc. Some examples disclosed herein selectively adjust sampling weights assigned to a panelist in view of the probabilities that the panelist is or is not watching television, is watching a certain program, etc. so as to identify a viewing population that can be used to calculate, for example, shares despite the uncertainties in the data.
0014Some examples disclosed herein compute variance or covariance metrics for analysis of viewing behavior across two or more television programs. Also, some disclosed examples can analyze ratings, shares, and/or other viewing metrics for a population subgroup or panelist of interest. For example, a demographic group can be analyzed with respect to what program the demographic group is watching or what portion of the demographic group is watching a particular program.
0015Examples disclosed herein more accurately identify ratings and/or shares with respect to uncertainties or probabilities in viewing behavior data and reduce errors in computing ratings and/or shares as compared to approaches that consider a limited range of probabilistic scenarios, only run a Monte Carlo simulation once, etc. Examples disclosed herein improve computational efficiency and reduce processing resources in considering the many scenarios that could arise for panelists or a group of panelists. Examples disclosed herein substantially eliminate the need to run a probabilistic scenario simulation hundreds or thousands of times. Rather, examples disclosed herein generate results that substantially approximate viewing metrics as if the simulations were performed an infinite number of times. Thus, disclosed examples provide a technical improvement in the field of ratings metrics over known methods that address uncertainties in viewing data in a limited fashion.
0016Although examples disclosed herein are discussed in the context of media viewing metrics such as television ratings and/or shares, examples disclosed herein can be utilized in other applications. For example, examples disclosed herein could be used for other types of media than television programs, such as radio. Also, examples disclosed herein could be used in applications other than media to analyze behavior of a population with respect to, for example, buying a product such as cereal.
0017<figref idref="DRAWINGS">FIG. 1</figref> illustrates an example system <b>100</b> for computing viewing metrics such as ratings and/or shares associated with one or more television programs. As illustrated in <figref idref="DRAWINGS">FIG. 1</figref>, a first household <b>102</b> includes a first panelist <b>104</b>. The first household <b>102</b> can include additional panelists. The first household <b>102</b> includes a first television <b>106</b>. A first panel meter <b>108</b> is communicatively coupled to the first television <b>106</b>. The first television <b>106</b> can be tuned to broadcast one or more channels <b>110</b><i>a</i>-<b>110</b><i>n</i>. Each of the channels <b>110</b><i>a</i>-<b>110</b><i>n </i>can provide one or more media or programs <b>112</b><i>a</i>-<b>112</b><i>n </i>to be viewed via the first television <b>106</b> by the first panelist <b>104</b>.
0018The first panel meter <b>108</b> collects data from the first television <b>106</b>, such as whether the first television <b>106</b> is turned on, to which of the channels <b>110</b><i>a</i>-<b>110</b><i>n </i>the first television is tuned, how long the first television <b>106</b> is tuned to the selected channel <b>110</b><i>a</i>-<b>110</b><i>n</i>, what time of day the first television <b>106</b> is tuned to the one of the channels <b>110</b><i>a</i>-<b>110</b><i>n</i>, etc. In the example of <figref idref="DRAWINGS">FIG. 1</figref>, the first panelist <b>104</b> is associated with a plurality of demographics <b>114</b>, such as age, gender, ethnicity, household size, etc. In some examples, the demographics <b>114</b> include data about the geographic location of the first household <b>102</b>, socioeconomic status, etc. In some examples, the first panel meter <b>108</b> collects and/or stores data about the demographics <b>114</b> the first panelist <b>104</b> (e.g., via one or more user inputs with respect to the demographics <b>114</b> of the first panelist <b>104</b>, census data, etc.).
0019The example system <b>100</b> includes a second household <b>116</b>. The second household <b>116</b> includes a second panelist <b>118</b>. The second household can include additional panelists. The second household <b>116</b> includes a second television <b>120</b> and a second panel meter <b>122</b> communicatively coupled to the second television <b>120</b>. The second panel meter <b>122</b> collects data from the second television <b>120</b> regarding, for example, which of the channels <b>110</b><i>a</i>-<b>110</b><i>n </i>the second television <b>120</b> is tuned to at a given time of day, and other data substantially as disclosed above in connection with the first television <b>106</b> and the first panel meter <b>108</b>. The second meter <b>122</b> can collect and/or store data about demographics <b>124</b> associated with the second panelist <b>118</b> (e.g., age, gender, etc. of the second panelist <b>118</b>).
0020The example system <b>100</b> can include other households in addition to the first household <b>102</b> and the second household <b>116</b> (e.g., n households <b>102</b>, <b>116</b>). Also any of the households <b>102</b>, <b>116</b> in the example system <b>100</b> of <figref idref="DRAWINGS">FIG. 1</figref> can include one or more panelists (e.g., n panelists <b>104</b>, <b>118</b>). Television viewing activity can be collected from any of the households in the example system <b>100</b> substantially as described herein with respect to the first and second households <b>102</b>, <b>116</b> of <figref idref="DRAWINGS">FIG. 1</figref>.
0021In the example system of <figref idref="DRAWINGS">FIG. 1</figref>, the first panel meter <b>108</b> and the second panel meter <b>122</b> are communicatively coupled to a processor <b>126</b> (e.g., via wireless connections) to transmit return path data to the processor <b>126</b>. The first panel meter <b>108</b> transmits a first data stream <b>128</b> to the processor <b>126</b> including viewing data for the first television <b>106</b> of the first household <b>102</b>. The second panel meter <b>122</b> transmits a second data stream <b>130</b> to the processor <b>126</b> including viewing data for the second television <b>120</b> of the second household <b>102</b>. The respective data streams <b>128</b>, <b>130</b> can include data such as the channel(s) <b>110</b><i>a</i>-<b>110</b><i>n </i>to which each television <b>106</b>, <b>120</b> is tuned at a certain time. The first and second data streams <b>128</b>, <b>130</b> can include demographic data about the panelists <b>104</b>, <b>118</b> of the respective households <b>102</b>, <b>116</b>. As disclosed below, the processor <b>126</b> stores the data streams <b>128</b>, <b>130</b> for analysis with respect to television viewing metrics.
0022In some examples, the first data stream <b>128</b> and/or the second data stream <b>130</b> includes data indicative of one or more uncertainties about the television viewing behavior of the first panelist <b>104</b> (or the first household <b>102</b>) and/or the second panelist <b>118</b> (or the second household <b>116</b>). For example, there may be uncertainty as to whether the first panelist <b>102</b> was co-viewing one of the programs <b>112</b><i>a</i>-<b>112</b><i>n </i>with another member of the first household <b>102</b>. As another example, there may have been a temporary technical error in the collection of data by the first and/or second panel meters <b>108</b>, <b>122</b> (e.g., an inability to collect data from the television(s) <b>106</b>, <b>120</b> for a period of time). Thus, at least a portion of the first data stream <b>128</b> and/or the second data stream <b>130</b> may include uncertain or probabilistic viewing activity data by the respective panelists <b>104</b>, <b>118</b> (and/or households <b>102</b>, <b>116</b>).
0023The example processor <b>126</b> of <figref idref="DRAWINGS">FIG. 1</figref> includes a viewing activity analyzer <b>132</b>. The example viewing activity analyzer <b>132</b> calculates viewing metrics such as ratings and/or shares for one or more of the programs <b>112</b><i>a</i>-<b>112</b><i>n </i>based on the data in the first data stream <b>128</b>, the second data stream <b>130</b>, and/or other data streams received from other households in the example system <b>100</b>. The example viewing activity analyzer <b>132</b> considers any uncertainties in the first data stream <b>128</b> and/or the second data stream <b>130</b> by calculating the viewing metrics using one or more algorithms that account for probabilities with respect to whether or not the panelist(s) <b>104</b>, <b>118</b> are watching television, what program each panelist <b>104</b>, <b>118</b> is watching, etc.
0024The example viewing activity analyzer <b>132</b> generates one or more viewing metric outputs <b>134</b>. The viewing metric output(s) <b>134</b> can include ratings and/or shares for one or more of the programs <b>112</b><i>a</i>-<b>112</b><i>n</i>. In some examples, the viewing metric output(s) <b>134</b> can include analysis results with respect to viewing activity of a population subgroup of interest, such as a particular demographic subgroup (e.g., an age group). The viewing metric output(s) <b>134</b> can be presented via one or more output devices <b>136</b>, such as a display screen of a personal computing device (e.g., associated with the processor <b>126</b>).
0025<figref idref="DRAWINGS">FIG. 2</figref> is a block diagram of an example implementation of the viewing activity analyzer <b>132</b> of <figref idref="DRAWINGS">FIG. 1</figref>. As illustrated in <figref idref="DRAWINGS">FIG. 2</figref>, the example viewing activity analyzer <b>132</b> includes a data collector <b>200</b>. The example data collector <b>200</b> receives one or more of the data streams <b>128</b>, <b>130</b> from the panel meters <b>108</b>, <b>122</b>. As disclosed above, the data streams <b>128</b>, <b>130</b> include data such as date, time, and/or duration that the television(s) <b>106</b>, <b>120</b> were turned on; the channel(s) <b>110</b><i>a</i>-<b>110</b><i>n </i>to which the television(s) <b>106</b>, <b>120</b> were tuned; the programs <b>112</b><i>a</i>-<b>112</b><i>n </i>broadcast by the channel(s) <b>110</b><i>a</i>-<b>110</b><i>n</i>; the respective demographics <b>114</b>, <b>124</b> of the panelists <b>104</b>, <b>118</b>, etc. In some examples, the data collector <b>200</b> filters and/or formats the data streams <b>128</b>, <b>130</b> for processing by the viewing activity analyzer <b>132</b>. The data streams <b>128</b>, <b>130</b> received by the data collector <b>200</b> are stored in a database <b>202</b> of the example viewing activity analyzer <b>132</b> of <figref idref="DRAWINGS">FIG. 2</figref>.
0026The example viewing activity analyzer <b>132</b> includes a sampling weight assigner <b>204</b>. The example sampling weight assigner <b>204</b> assigns a sampling weight <b>205</b> to each panelist <b>104</b>, <b>118</b> based on, for example, the respective demographics <b>114</b>, <b>124</b> of each panelist <b>104</b>, <b>118</b>. The sampling weight(s) <b>205</b> assigned by the example weight assigner <b>204</b> to each panelist <b>104</b>, <b>118</b> is indicative of a number of other television viewers that each panelist <b>104</b>, <b>118</b> represents based on, for example, one or more similar demographics <b>114</b>, <b>124</b> (e.g., age, gender, socioeconomic status). For example, if the sampling weight assigner <b>204</b> assigns a sampling weight <b>205</b> having a value of ten to the first panelist <b>104</b>, the first panelist <b>104</b> represents ten people sharing similar demographics <b>114</b> as the first panelist <b>104</b>.
0027In some examples, the sampling weight <b>205</b> is based on whether the panel meter(s) <b>108</b>, <b>122</b> were working properly during a time period in which the data of the data stream(s) <b>128</b>, <b>130</b> was collected. For example, if a known power outage affected the first household <b>102</b> and, thus, the ability of the first panelist <b>104</b> to watch the first television <b>106</b> and the first panel meter <b>108</b> to collect data, the example sampling weight assigner <b>204</b> can adjust the sampling weight <b>205</b> assigned to the first panelist <b>104</b> to reflect a number of people who were affected by the power outage.
0028In the example of <figref idref="DRAWINGS">FIG. 2</figref>, the sampling weight(s) <b>205</b> assigned to the panelist(s) <b>104</b>, <b>118</b> can be based on one or more sampling weight rule(s) <b>206</b> stored in the database <b>202</b> of <figref idref="DRAWINGS">FIG. 2</figref>. The sampling weight rules <b>206</b> can include one or more rules with respect to a value of the sampling weight(s) <b>205</b> to be assigned to each panelist <b>104</b>, <b>118</b> based on demographic factors such as age, gender, household size, etc. The example sampling weight assigner <b>204</b> of <figref idref="DRAWINGS">FIG. 2</figref> compares the demographic data <b>114</b>, <b>124</b> in the data streams <b>128</b>, <b>130</b> to the sampling weight rule(s) <b>206</b> to determine the sampling weight(s) <b>205</b> to assign to the panelist(s) <b>104</b>, <b>118</b>.
0029The example viewing activity analyzer <b>132</b> of <figref idref="DRAWINGS">FIG. 2</figref> includes a probability identifier <b>208</b>. The example probability identifier <b>208</b> analyzes the first and second data streams <b>128</b>, <b>130</b> to identify any uncertainties in the data streams <b>128</b>, <b>130</b>. For example, the probability identifier <b>208</b> can identify missing data in the first and/or second data streams <b>128</b>, <b>130</b> with respect to, for example, data regarding whether or not the panelist(s) <b>104</b>, <b>118</b> where watching the television(s) <b>106</b>, <b>120</b>, what program(s) <b>112</b><i>a</i>-<b>112</b><i>n </i>the panelist(s) <b>104</b>, <b>118</b> were watching, etc. The probability identifier <b>208</b> can identify inconsistencies in the first and/or second data streams <b>128</b> such as data corresponding to one or more program(s) <b>112</b><i>a</i>-<b>112</b><i>n </i>that did not air during the time period for which the data was collected. The probability identifier <b>208</b> can identify potential co-viewing activity based on, for example, a number of panelists <b>104</b>, <b>118</b> associated with each household <b>102</b>, <b>116</b>.
0030In other examples, the probability identifier <b>208</b> does not identify any uncertainties in the first and/or second data streams <b>128</b>, <b>130</b>. For example, the data stream(s) <b>128</b>, <b>130</b> can include data with respect to the television program(s) <b>112</b><i>a</i>-<b>112</b><i>n </i>that the panelist(s) <b>104</b>, <b>118</b> were watching that has not been affected by, for example, any technical errors in the data collection.
0031In some examples, the probability identifier <b>208</b> assigns one or more viewing probabilities <b>209</b> to the panelist(s) <b>104</b>, <b>118</b> based on the uncertainties identified in the data stream(s) <b>128</b>, <b>130</b> with respect to, for example, whether or not the panelist(s) <b>104</b>, <b>118</b> are watching television, what program(s) <b>112</b><i>a</i>-<b>112</b><i>n </i>the panelist(s) <b>104</b>, <b>118</b> could have watched, etc. The example probability identifier <b>208</b> assigns the probabilities <b>209</b> based on one or more probability rules <b>207</b> stored in the example database <b>202</b> of <figref idref="DRAWINGS">FIG. 2</figref>. The probability rules <b>207</b> can include predefined rules with respect to probability values to be assigned to the panelists <b>104</b>, <b>118</b> based on, for example, the number of programs <b>112</b><i>a</i>-<b>112</b><i>n </i>that the panelist(s) <b>104</b>, <b>118</b> could be watching at a given time, the sampling weights <b>205</b> assigned to the panelist(s) <b>104</b>, <b>118</b>, historical viewing data for the respective panelist(s) <b>104</b>, <b>118</b> stored in the database <b>202</b>, etc.
0032Table 1, below, is an example table generated by the example probability identifier <b>208</b> of <figref idref="DRAWINGS">FIG. 2</figref>. Table 1 includes probabilistic viewing activity for a plurality of panelists (e.g., the panelists <b>104</b>, <b>118</b> of <figref idref="DRAWINGS">FIG. 1</figref>) whose viewing data is received by the data collector <b>200</b> of the example viewing activity analyzer <b>132</b>. Table 1 includes the probabilistic viewing activity with respect to a first program <b>112</b><i>a </i>(P<sub>112a</sub>), a second program <b>112</b><i>b </i>(P<sub>112b</sub>), and a third first program <b>112</b><i>c</i>(P<sub>112c</sub>). Table 1 also includes probabilities with respect to whether or not the panelists are watching television (P<sub>0</sub>).
0033<tables id="TABLE-US-00001" num="00001"><table frame="none" colsep="0" rowsep="0"><tgroup align="left" colsep="0" rowsep="0" cols="1"><colspec colname="1" colwidth="217pt" align="center" /><thead><row><entry namest="1" nameend="1" rowsep="1">TABLE 1</entry></row></thead><tbody valign="top"><row><entry namest="1" nameend="1" align="center" rowsep="1" /></row><row><entry>Viewing Activity Probabilities</entry></row></tbody></tgroup><tgroup align="left" colsep="0" rowsep="0" cols="7"><colspec colname="1" colwidth="49pt" align="left" /><colspec colname="2" colwidth="49pt" align="left" /><colspec colname="3" colwidth="35pt" align="center" /><colspec colname="4" colwidth="21pt" align="center" /><colspec colname="5" colwidth="21pt" align="center" /><colspec colname="6" colwidth="21pt" align="center" /><colspec colname="7" colwidth="21pt" align="center" /><tbody valign="top"><row><entry /><entry>Age (e.g.,</entry><entry>Sampling</entry><entry /><entry /><entry /><entry /></row><row><entry /><entry>demographics</entry><entry>Weights</entry></row><row><entry>Panelist</entry><entry>114, 124)</entry><entry>(205)</entry><entry>P<sub>0</sub></entry><entry>P<sub>112a</sub></entry><entry>P<sub>112b</sub></entry><entry>P<sub>112c</sub></entry></row><row><entry namest="1" nameend="7" align="center" rowsep="1" /></row></tbody></tgroup><tgroup align="left" colsep="0" rowsep="0" cols="7"><colspec colname="1" colwidth="49pt" align="left" /><colspec colname="2" colwidth="49pt" align="left" /><colspec colname="3" colwidth="35pt" align="center" /><colspec colname="4" colwidth="21pt" align="char" char="." /><colspec colname="5" colwidth="21pt" align="char" char="." /><colspec colname="6" colwidth="21pt" align="char" char="." /><colspec colname="7" colwidth="21pt" align="char" char="." /><tbody valign="top"><row><entry>A (e.g.,</entry><entry>Young</entry><entry>10</entry><entry>0.5</entry><entry>.5</entry><entry>0</entry><entry>0</entry></row><row><entry>panelist 104)</entry></row><row><entry>B (e.g.,</entry><entry>Young</entry><entry>60</entry><entry>0</entry><entry>.33</entry><entry>.33</entry><entry>.33</entry></row><row><entry>panelist 118)</entry></row><row><entry>C</entry><entry>Young</entry><entry>20</entry><entry>1</entry><entry>0</entry><entry>0</entry><entry>0</entry></row><row><entry>D</entry><entry>Middle</entry><entry>80</entry><entry>.1</entry><entry>.2</entry><entry>.3</entry><entry>.4</entry></row><row><entry>E</entry><entry>Middle</entry><entry>40</entry><entry>0</entry><entry>0</entry><entry>1</entry><entry>0</entry></row><row><entry>F</entry><entry>Middle</entry><entry>70</entry><entry>0</entry><entry>.3</entry><entry>.5</entry><entry>.2</entry></row><row><entry>G</entry><entry>Old</entry><entry>90</entry><entry>.25</entry><entry>.25</entry><entry>.25</entry><entry>.25</entry></row><row><entry>H</entry><entry>Old</entry><entry>30</entry><entry>.4</entry><entry>.3</entry><entry>.2</entry><entry>.1</entry></row><row><entry>I</entry><entry>Old</entry><entry>50</entry><entry>.1</entry><entry>.7</entry><entry>0</entry><entry>.2</entry></row><row><entry namest="1" nameend="7" align="center" rowsep="1" /></row></tbody></tgroup></table></tables>
0034As illustrated above, the example Table 1 includes panelist identifiers (e.g., letters A-H), associated demographics (e.g., age), and respective sampling weights <b>205</b> assigned to the panelists (e.g., by the sampling weight assigner <b>204</b> of the viewing activity analyzer <b>132</b>). In example Table 1, the values in the third column P<sub>0 </sub>represent a probability that a respective panelist is not watching television, the values in the fourth column P<sub>112a </sub>represent a probability that a panelist is watching the first program <b>112</b><i>a</i>, the values in the fifth column P<sub>112b </sub>represent a probability that a panelist is watching the second program <b>112</b><i>b</i>, and the value in the sixth column P<sub>112c </sub>represent a probability that a panelist is watching the third program <b>112</b><i>c. </i>
0035For example, referring to Table 1, the probability identifier <b>208</b> determines based on the first data stream <b>128</b> that Panelist A (e.g., the first panelist <b>104</b> of <figref idref="DRAWINGS">FIG. 1</figref>) is either not watching television with a 50% probability or watching the first program <b>112</b><i>a </i>with a 50% probability.
0036As another example, the probability identifier <b>208</b> determines based on, for example, the second data stream <b>130</b>, that Panelist B (e.g., the second panelist <b>118</b> of <figref idref="DRAWINGS">FIG. 1</figref>) is watching television. However, the probability identifier <b>208</b> is unable to determine which program <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c </i>Panelist B is watching based on the second data stream <b>130</b>. Accordingly, the probability identifier <b>208</b> assigns equal probabilities to Panelist B with respect to the first, second, and third programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c</i>. As another example, the probability identifier <b>208</b> can determine that Panelist E is watching the second program <b>112</b><i>b </i>based on a data stream received by the data collector <b>200</b> for Panelist E. Accordingly, the probability identifier <b>208</b> assigns Panelist E a probability of “1” based on the data indicating that Panelist E is watching the first program <b>112</b><i>a</i>. Thus, as disclosed above, the example probability identifier <b>208</b> analyzes the data streams (e.g., the data streams <b>128</b>, <b>130</b>) and assigns probabilities <b>209</b> with respect to viewing activity based on the data, including any uncertainties in the data.
0037The example viewing activity analyzer <b>132</b> of <figref idref="DRAWINGS">FIG. 2</figref> includes a ratings calculator <b>210</b>. The example ratings calculator <b>210</b> calculates one or more ratings <b>211</b> for one or more of the programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c </i>based on the data in the data streams (e.g., the data streams <b>128</b>, <b>130</b>) in view of the probabilities <b>209</b> determined by the probability identifier <b>208</b> (e.g., as provided in Table 1). In the example of <figref idref="DRAWINGS">FIG. 2</figref>, the ratings calculator <b>210</b> also determines a null rating <b>211</b> representative of a percent of panelists not watching television (e.g., F<sub>0</sub>). The example ratings calculator <b>210</b> of <figref idref="DRAWINGS">FIG. 2</figref> employs a plurality of algorithms that account for the sampling weights <b>205</b> assigned to the respective panelists by the sampling weight assigner <b>204</b> and the probabilities <b>209</b> assigned by the probability identifier <b>208</b>.
0038For example, the ratings calculator <b>210</b> can apply the following equations to determine the expected ratings <b>211</b> for the first, second, and third programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c </i>and the percent of televisions not tuned to any of the programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c: </i>
0039Where p<sub>k,i </sub>is a probability that the k<sup>th </sup>panelist is watching the i<sup>th </sup>program, w<sub>k </sub>is a sampling weight associated with the k<sup>th </sup>panelist, and n is the number of panelists,
0040<maths id="MATH-US-00001" num="00001"><math overflow="scroll"><mtable><mtr><mtd><mrow><mrow><mi>E</mi><mo></mo><mrow><mo>[</mo><msub><mi>R</mi><mi>i</mi></msub><mo>]</mo></mrow></mrow><mo>=</mo><mfrac><mrow><munderover><mo>∑</mo><mrow><mi>k</mi><mo>=</mo><mn>1</mn></mrow><mi>n</mi></munderover><mo></mo><mrow><msub><mi>w</mi><mi>k</mi></msub><mo></mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>i</mi></mrow></msub></mrow></mrow><mrow><munderover><mo>∑</mo><mrow><mi>k</mi><mo>=</mo><mn>1</mn></mrow><mi>n</mi></munderover><mo></mo><msub><mi>w</mi><mi>k</mi></msub></mrow></mfrac></mrow></mtd><mtd><mrow><mo>(</mo><mn>1</mn><mo>)</mo></mrow></mtd></mtr><mtr><mtd><mrow><mrow><mi>Var</mi><mo></mo><mrow><mo>[</mo><msub><mi>R</mi><mi>i</mi></msub><mo>]</mo></mrow></mrow><mo>=</mo><mfrac><mrow><munderover><mo>∑</mo><mrow><mi>k</mi><mo>=</mo><mn>1</mn></mrow><mi>n</mi></munderover><mo></mo><mrow><mrow><msubsup><mi>w</mi><mi>k</mi><mn>2</mn></msubsup><mo></mo><mrow><mo>(</mo><mrow><mn>1</mn><mo>-</mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>i</mi></mrow></msub></mrow><mo>)</mo></mrow></mrow><mo></mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>i</mi></mrow></msub></mrow></mrow><msup><mrow><mo>(</mo><mrow><munderover><mo>∑</mo><mrow><mi>k</mi><mo>=</mo><mn>1</mn></mrow><mi>n</mi></munderover><mo></mo><msub><mi>w</mi><mi>k</mi></msub></mrow><mo>)</mo></mrow><mn>2</mn></msup></mfrac></mrow></mtd><mtd><mrow><mo>(</mo><mn>2</mn><mo>)</mo></mrow></mtd></mtr><mtr><mtd><mrow><mrow><mi>Cov</mi><mo></mo><mrow><mo>[</mo><mrow><msub><mi>R</mi><mi>i</mi></msub><mo>,</mo><msub><mi>R</mi><mi>j</mi></msub></mrow><mo>]</mo></mrow></mrow><mo>=</mo><mrow><mo>-</mo><mfrac><mrow><munderover><mo>∑</mo><mrow><mi>k</mi><mo>=</mo><mn>1</mn></mrow><mi>n</mi></munderover><mo></mo><mrow><msubsup><mi>w</mi><mi>k</mi><mn>2</mn></msubsup><mo></mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>i</mi></mrow></msub><mo></mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>j</mi></mrow></msub></mrow></mrow><msup><mrow><mo>(</mo><mrow><munderover><mo>∑</mo><mrow><mi>k</mi><mo>=</mo><mn>1</mn></mrow><mi>n</mi></munderover><mo></mo><msub><mi>w</mi><mi>k</mi></msub></mrow><mo>)</mo></mrow><mn>2</mn></msup></mfrac></mrow></mrow></mtd><mtd><mrow><mo>(</mo><mn>3</mn><mo>)</mo></mrow></mtd></mtr></mtable></math></maths><img file="US10791355B2_D0001.tif" />
0041Thus, expected ratings, variance, and covariance calculations are summed across the number of panelists n (e.g., the Panelists A-H of Table 1, above). In some examples, the ratings calculator <b>210</b> utilizes a normalized sampling weight or weighted average v<sub>k </sub>for the sampling weights <b>205</b> associated with the panelists, where
0042<maths id="MATH-US-00002" num="00002"><math overflow="scroll"><mrow><msub><mi>v</mi><mi>k</mi></msub><mo>=</mo><mrow><mfrac><msub><mi>w</mi><mi>k</mi></msub><mrow><munderover><mo>∑</mo><mrow><mi>k</mi><mo>=</mo><mn>1</mn></mrow><mi>n</mi></munderover><mo></mo><msub><mi>w</mi><mi>k</mi></msub></mrow></mfrac><mo>.</mo></mrow></mrow></math></maths><img file="US10791355B2_D0002.tif" /><br /> Equations 1-3 above can be modified to include the normalized weight v<sub>k </sub>as follows: <br /><i>E[R</i><sub>i</sub>]=Σ<sub>k=1</sub><sup>n</sup><i>v</i><sub>k</sub><i>p</i><sub>k,i</sub> (4)<br />Var[<i>R</i><sub>i</sub>]=Σ<sub>k=1</sub><sup>n</sup><i>v</i><sub>k</sub><sup>2</sup>(1−<i>p</i><sub>k,i</sub>)<i>p</i><sub>k,i</sub> (5)<br />Cov[<i>R</i><sub>i</sub><i>,R</i><sub>j</sub>]=−Σ<sub>k=1</sub><sup>n</sup><i>v</i><sub>k</sub><sup>2</sup><i>p</i><sub>k,i</sub><i>p</i><sub>k,j</sub> (6)
0043In Equation (4), above, the expected ratings <b>211</b> for the i<sup>th </sup>program (e.g., one of programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c</i>) are determined by summing the weighted average v<sub>k </sub>by the probability that the panelists (and, thus, the number of people each panelist represents) are watching the i<sup>th </sup>program.
0044In Equation (5), above, the variance calculation accounts for a probability that, for example, Panelist A (e.g., the first panelist <b>104</b> of <figref idref="DRAWINGS">FIG. 1</figref>) is not watching television and a probability that Panelist A is watching the i<sup>th </sup>program (e.g., the first program <b>112</b><i>a</i>). Thus, Equation (5) accounts for uncertainties in the first data stream <b>128</b> with respect to whether or not the first panelist <b>104</b> is watching the first program <b>112</b><i>a </i>(e.g., p<sub>k,i</sub>) or is not watching television (e.g., 1−p<sub>k,i</sub>)) by considering both probabilities.
0045Equation (5) also considers the sampling weight <b>205</b> assigned to Panelist A (e.g., the first panelist <b>104</b>) and, accordingly, a portion of the population represented by the Panelist A. For example, as indicated in example Table 1, above, the Panelist A is assigned a weight of ten. Thus, Panelist A represents ten individuals sharing, for example, a similar age demographic as Panelist A. As such, if there is a 20% probability that Panelist A is watching the first program <b>112</b><i>a</i>, then the ten people represented by the Panelist A are also considered to be watching the first program <b>112</b><i>a </i>with a probability of 20%. Thus, Equation (5) considers the probability that Panelist A is watching television and/or is watching one of the programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c </i>as well as the portion of the population represented by the first panelist <b>104</b>. In Equation (5), the variance is summed across the panelists to account for the fact that different panelists are associated with different probabilities of viewing a program and/or different probabilities with respect to not viewing television.
0046Referring to Table 1 above including the probabilities <b>209</b> of television viewership activity, the example ratings calculator <b>210</b> calculates the ratings <b>211</b> for the first program <b>112</b><i>a</i>, the second program <b>112</b><i>b</i>, and the third program <b>112</b><i>c </i>using Equations (1) or (4). The ratings calculator <b>210</b> also calculates a null rating <b>211</b> representing a percentage of panelists not watching television. For example, the ratings calculator <b>210</b> can calculate the following expected ratings <b>211</b> for P<sub>0</sub>, P<sub>112a</sub>, P<sub>112b</sub>, P<sub>112c </sub>of Table 1 as follows: <br /><i>E[R</i><sub>i</sub>]=[0.1611 0.2855 0.3274 0.2259] (7)
0047Also, the example ratings calculator <b>210</b> can calculate a covariance matrix σ<sup>2</sup>(R<sub>i</sub>,R<sub>j</sub>) based on the variance equations (e.g., Equations (2) or (5)) and the covariance equations (e.g., Equations (3) or (6)) as follows:
0048<maths id="MATH-US-00003" num="00003"><math overflow="scroll"><mtable><mtr><mtd><mrow><mrow><msup><mi>σ</mi><mn>2</mn></msup><mo></mo><mrow><mo>(</mo><mrow><msub><mi>R</mi><mi>i</mi></msub><mo></mo><msub><mi>R</mi><mi>j</mi></msub></mrow><mo>)</mo></mrow></mrow><mo>=</mo><mrow><mo>(</mo><mtable><mtr><mtd><mrow><mo>+</mo><mn>0.0126</mn></mrow></mtd><mtd><mrow><mo>-</mo><mn>0.0047</mn></mrow></mtd><mtd><mrow><mo>-</mo><mn>0.0038</mn></mrow></mtd><mtd><mrow><mo>-</mo><mn>0.0042</mn></mrow></mtd></mtr><mtr><mtd><mrow><mo>-</mo><mn>0.0047</mn></mrow></mtd><mtd><mrow><mo>+</mo><mn>0.0253</mn></mrow></mtd><mtd><mrow><mo>-</mo><mn>0.0103</mn></mrow></mtd><mtd><mrow><mo>-</mo><mn>0.0103</mn></mrow></mtd></mtr><mtr><mtd><mrow><mo>-</mo><mn>0.0038</mn></mrow></mtd><mtd><mrow><mo>-</mo><mn>0.0103</mn></mrow></mtd><mtd><mrow><mo>+</mo><mn>0.0248</mn></mrow></mtd><mtd><mrow><mo>-</mo><mn>0.0108</mn></mrow></mtd></mtr><mtr><mtd><mrow><mo>-</mo><mn>0.0042</mn></mrow></mtd><mtd><mrow><mo>-</mo><mn>0.0103</mn></mrow></mtd><mtd><mrow><mo>-</mo><mn>0.0108</mn></mrow></mtd><mtd><mrow><mo>+</mo><mn>0.0253</mn></mrow></mtd></mtr></mtable><mo>)</mo></mrow></mrow></mtd><mtd><mrow><mo>(</mo><mn>8</mn><mo>)</mo></mrow></mtd></mtr></mtable></math></maths><img file="US10791355B2_D0003.tif" />
0049The covariance matrix (8) indicates relationships between, for example, the first program <b>112</b><i>a </i>and the other programs <b>112</b><i>b</i>, <b>112</b><i>c</i>. In the example covariance matrix (8), the diagonals of the matrix are computed by the ratings calculator <b>210</b> based on the variance (e.g., Equations (2) or (5)) and the off-diagonals of the matrix are computed based on the covariance (e.g., Equations (3) or (6)). In the example covariance matrix (8), the off-diagonals include negative values. The negative values of the off-diagonals in the covariance matrix (8) reflect the fact out of the potential viewing population, more people in the population who are watching one program (e.g., the first program <b>112</b><i>a</i>) means that less people in the population are able to watch the other programs (e.g., the second program <b>112</b><i>b</i>, the third program <b>112</b><i>c</i>). Also, the ratings calculator <b>210</b> considers the population that may not be watching television because that population is a part of the total potential viewing population. Thus, the example ratings calculator <b>210</b> of <figref idref="DRAWINGS">FIG. 2</figref> calculates the ratings for the first, second, and third programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c </i>based the probabilities that the panelists (and, thus, the portion of the population they represent) are viewing television or not viewing television.
0050The example viewing activity analyzer <b>132</b> of <figref idref="DRAWINGS">FIG. 2</figref> includes a share calculator <b>212</b>. The share calculator <b>212</b> determines share(s) <b>213</b>, or a percentage of televisions that are in use that are tuned to a certain program. The shares computed by the example share calculator <b>212</b> are conditional based on the panelists (e.g., the panelists <b>104</b>, <b>118</b> of <figref idref="DRAWINGS">FIG. 1</figref>) who are watching television. As disclosed above, there may be uncertainties with respect to whether a panelist such as the first panelist <b>104</b> and/or the second panelist <b>118</b> is watching television. Thus, the number of panelists who watching are television is a random variable. The shares calculator <b>212</b> considers the different panelists who may be watching television, as each panelist's sampling weight <b>205</b> may differ from another panelist.
0051The random variables with respect to the number of panelists who are watching television and the different sampling weights <b>205</b> associated each panelist can consume extensive resources of a processor (e.g., the processor <b>126</b> of <figref idref="DRAWINGS">FIG. 1</figref>) to calculate exact shares values. For example, to calculate the exact shares values, the shares calculator <b>212</b> would need to run multiple simulations considering all of the programs <b>112</b><i>a</i>-<b>112</b><i>n </i>the panelists could be watching, with different panelists treated as watching different programs for each simulation. The multiple simulations consume resources of the processor <b>126</b>, which can increase a time to perform the analysis and decrease efficiency. However, the shares calculator <b>212</b> of <figref idref="DRAWINGS">FIG. 2</figref> increases the efficiency in determining the share(s) <b>213</b> by approximating the share(s) <b>213</b> as a conditional distribution of ratings. In the example of <figref idref="DRAWINGS">FIG. 2</figref>, the calculation of the share(s) <b>213</b> based on the conditional distribution of ratings converges to the exact shares values as the number of panelists considered increases. For a large panel size (e.g., thousands of panelists), the difference between the shares <b>213</b> calculated by the shares calculator <b>212</b> based on the conditional distribution and the exact shares values (e.g., calculated based on multiple simulations with the panelists watching different programs in each simulation) is substantially negligible.
0052The example shares calculator <b>212</b> of <figref idref="DRAWINGS">FIG. 2</figref> determines a probability that a panelist is watching a particular program (e.g., the first, second, or third programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c</i>), on the condition that the panelist is watching television. The shares calculator <b>212</b> calculates a share weight <b>215</b> for each panelist (e.g., the panelists <b>104</b>, <b>118</b> of <figref idref="DRAWINGS">FIG. 1</figref>) based on a product of the sampling weight <b>205</b> assigned to the panelist and a probability that the panelist is not viewing television. The shares calculator <b>212</b> calculates a normalized share weight z<sub>k </sub>based on the sampling weights <b>205</b> (e.g., the sampling weights <b>205</b> in Table 1) as follows:
0053<maths id="MATH-US-00004" num="00004"><math overflow="scroll"><mtable><mtr><mtd><mrow><msub><mi>z</mi><mi>k</mi></msub><mo>=</mo><mfrac><mrow><msub><mi>w</mi><mi>k</mi></msub><mo></mo><mrow><mo>(</mo><mrow><mn>1</mn><mo>-</mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mn>0</mn></mrow></msub></mrow><mo>)</mo></mrow></mrow><mrow><munderover><mo>∑</mo><mrow><mi>k</mi><mo>=</mo><mn>1</mn></mrow><mi>n</mi></munderover><mo></mo><mrow><msub><mi>w</mi><mi>k</mi></msub><mo></mo><mrow><mo>(</mo><mrow><mn>1</mn><mo>-</mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mn>0</mn></mrow></msub></mrow><mo>)</mo></mrow></mrow></mrow></mfrac></mrow></mtd><mtd><mrow><mo>(</mo><mn>9</mn><mo>)</mo></mrow></mtd></mtr></mtable></math></maths><img file="US10791355B2_D0004.tif" />
0054Equation (9) adjusts the respective sampling weights <b>205</b> assigned to the panelists based on the probabilities p<sub>k,0 </sub>that the panelists are not watching television. The shares calculator <b>212</b> calculates a conditional share probability that if a panelist is watching television, then the panelist is watching the i<sup>th </sup>program, as follows:
0055<maths id="MATH-US-00005" num="00005"><math overflow="scroll"><mtable><mtr><mtd><mrow><msub><mi>s</mi><mrow><mi>k</mi><mo>,</mo><mi>i</mi></mrow></msub><mo>=</mo><mfrac><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>i</mi></mrow></msub><mrow><mn>1</mn><mo>-</mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mn>0</mn></mrow></msub></mrow></mfrac></mrow></mtd><mtd><mrow><mo>(</mo><mn>10</mn><mo>)</mo></mrow></mtd></mtr></mtable></math></maths><img file="US10791355B2_D0005.tif" />
0056In the example of <figref idref="DRAWINGS">FIG. 2</figref>, the shares calculator <b>212</b> generates a table including share weights <b>215</b> for each panelist in Table 1 (above) and conditional probabilities with respect to whether each panelist watching the first, second, or third programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c </i>Table 2, below, is an example table generated by the shares calculator <b>212</b> based on Equations (9) and (10) for the panelists in Table 1:
0057<tables id="TABLE-US-00002" num="00002"><table frame="none" colsep="0" rowsep="0"><tgroup align="left" colsep="0" rowsep="0" cols="1"><colspec colname="1" colwidth="217pt" align="center" /><thead><row><entry namest="1" nameend="1" rowsep="1">TABLE 2</entry></row></thead><tbody valign="top"><row><entry namest="1" nameend="1" align="center" rowsep="1" /></row><row><entry>Conditional Share Probabilities</entry></row></tbody></tgroup><tgroup align="left" colsep="0" rowsep="0" cols="6"><colspec colname="1" colwidth="49pt" align="left" /><colspec colname="2" colwidth="49pt" align="left" /><colspec colname="3" colwidth="35pt" align="center" /><colspec colname="4" colwidth="28pt" align="center" /><colspec colname="5" colwidth="28pt" align="center" /><colspec colname="6" colwidth="28pt" align="center" /><tbody valign="top"><row><entry /><entry>Age (e.g.,</entry><entry>Share</entry><entry /><entry /><entry /></row><row><entry /><entry>demographics</entry><entry>Weights</entry></row><row><entry>Panelist</entry><entry>114, 124)</entry><entry>(215)</entry><entry>S<sub>1</sub></entry><entry>S<sub>2</sub></entry><entry>S<sub>3</sub></entry></row><row><entry namest="1" nameend="6" align="center" rowsep="1" /></row></tbody></tgroup><tgroup align="left" colsep="0" rowsep="0" cols="6"><colspec colname="1" colwidth="49pt" align="left" /><colspec colname="2" colwidth="49pt" align="left" /><colspec colname="3" colwidth="35pt" align="char" char="." /><colspec colname="4" colwidth="28pt" align="char" char="." /><colspec colname="5" colwidth="28pt" align="char" char="." /><colspec colname="6" colwidth="28pt" align="char" char="." /><tbody valign="top"><row><entry>A (e.g.,</entry><entry>Young</entry><entry>5</entry><entry>1</entry><entry>0</entry><entry>0</entry></row><row><entry>panelist 104)</entry></row><row><entry>B (e.g.,</entry><entry>Young</entry><entry>60</entry><entry>.333</entry><entry>.333</entry><entry>.333</entry></row><row><entry>panelist 118)</entry></row><row><entry>C</entry><entry>Young</entry><entry>0</entry><entry>N/A</entry><entry>N/A</entry><entry>N/A</entry></row><row><entry>D</entry><entry>Middle</entry><entry>72</entry><entry>.222</entry><entry>.333</entry><entry>.444</entry></row><row><entry>E</entry><entry>Middle</entry><entry>40</entry><entry>0</entry><entry>1</entry><entry>0</entry></row><row><entry>F</entry><entry>Middle</entry><entry>70</entry><entry>.3</entry><entry>.5</entry><entry>.2</entry></row><row><entry>G</entry><entry>Old</entry><entry>67.5</entry><entry>.333</entry><entry>.333</entry><entry>.333</entry></row><row><entry>H</entry><entry>Old</entry><entry>18</entry><entry>.5</entry><entry>.333</entry><entry>.166</entry></row><row><entry>I</entry><entry>Old</entry><entry>45</entry><entry>.777</entry><entry>0</entry><entry>.222</entry></row><row><entry namest="1" nameend="6" align="center" rowsep="1" /></row></tbody></tgroup></table></tables>
0058For example, in Table 1, above, Panelist C is assigned a sampling weight <b>205</b> of 20 by the example weight assigner <b>204</b> of <figref idref="DRAWINGS">FIG. 2</figref> (e.g., based on demographics associated with Panelist C). However, Panelist C is also assigned a value of 1 with respect to P<sub>0</sub>, indicating that Panelist C is not watching television. Accordingly, the example shares calculator <b>212</b> adjusts the sampling weight <b>205</b> assigned to Panelist C such that Panelist C has a share weight <b>215</b> of 0 because Panelist C is not watching television. As illustrated in example Table 2, because Panelist C is not watching television, Panelist C does not contribute to the calculation of the shares <b>213</b> for the first, second, and/or third programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c. </i>
0059As another example, in Table 1, Panelist A is assigned a 50% probability of not watching television and a 50% probability of watching the first program <b>112</b><i>a</i>. Accordingly, the example shares calculator <b>212</b> adjusts the share weight <b>215</b> assigned to Panelist A in Table 2. Also, the example shares calculator <b>212</b> determines a conditional share probability indicating that if Panelist A is watching a program, then Panelist A is watching the first program <b>112</b><i>a </i>(e.g., as indicated by the value “1” for S<sub>1</sub>).
0060As another example, Table 1 indicates that there is a 10% probability that Panelist D is not watching television. Thus, there is a 90% probability that Panelist D is watching television. The example shares calculator <b>212</b> of <figref idref="DRAWINGS">FIG. 2</figref> adjusts the sampling weight <b>205</b> (e.g., 80) assigned to Panelist D to obtain the share weight <b>215</b> for Panelist D (e.g., 80*0.9=72). Thus, although for ratings purposes, Panelist D represents 80 people, for purposes of determining shares, Panelist D represents 72 people. As also indicated in Table 1, there is a 20% probability that Panelist D is watching the first program <b>112</b><i>a</i>, a 30% probability of watching the second program <b>112</b><i>b</i>, and a 40% probability of watching the third program <b>112</b><i>c</i>, and, thus, a 90% probability that Panelist D is watching one of the programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c</i>. The shares calculator <b>212</b> calculates, for example, a conditional share probability S<sub>1 </sub>that Panelist D is watching the first program <b>112</b><i>a </i>based on the probability that Panelist D is watching the first program from Table 1 (e.g., 0.2/0.9=0.222). The shares calculator <b>212</b> calculates a conditional share probability S<sub>2 </sub>that Panelist D is watching the second program <b>112</b><i>b </i>(e.g., 0.3/0.9=0.333) and a conditional share probability S<sub>3 </sub>that Panelist D is watching the third program <b>112</b><i>b </i>(e.g., 0.4/0.9=0.444). Thus, the conditional share probabilities S<sub>1</sub>, S<sub>2</sub>, S<sub>3 </sub>in Table 2 are based on the condition that a panelist is viewing television.
0061The examples shares calculator <b>212</b> of <figref idref="DRAWINGS">FIG. 2</figref> computes the expected shares for the ith program (e.g., first, second, and/or third programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c</i>), the variance, and covariance based on the data in example Table 2 as follows: <br /><i>E[S</i><sub>i</sub>]=Σ<sub>k=1</sub><sup>n</sup><i>z</i><sub>k</sub><i>s</i><sub>k,i</sub> (11)<br />Var[<i>S</i><sub>i</sub>]=Σ<sub>k=1</sub><sup>n</sup><i>z</i><sub>k</sub><sup>2</sup>(1−<i>s</i><sub>k,i</sub>)<i>s</i><sub>k,i</sub> (12)<br />Cov[<i>R</i><sub>i</sub><i>,R</i><sub>j</sub>]=−Σ<sub>k=1</sub><sup>n</sup><i>z</i><sub>k</sub><sup>2</sup><i>s</i><sub>k,i</sub><i>s</i><sub>k,j</sub> (13)
0062In some examples, if the ratings for the ith program (e.g., first, second, and/or third programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c</i>) have been calculated (e.g., as disclosed above with respect to Equations (1) or (4)), the shares calculator <b>212</b> calculates the expected shares as follows:
0063<maths id="MATH-US-00006" num="00006"><math overflow="scroll"><mtable><mtr><mtd><mrow><mrow><mi>E</mi><mo></mo><mrow><mo>[</mo><msub><mi>S</mi><mi>i</mi></msub><mo>]</mo></mrow></mrow><mo>=</mo><mfrac><mrow><mi>E</mi><mo></mo><mrow><mo>[</mo><msub><mi>R</mi><mi>i</mi></msub><mo>]</mo></mrow></mrow><mrow><mn>1</mn><mo>-</mo><mrow><mi>E</mi><mo></mo><mrow><mo>[</mo><msub><mi>R</mi><mn>0</mn></msub><mo>]</mo></mrow></mrow></mrow></mfrac></mrow></mtd><mtd><mrow><mo>(</mo><mn>14</mn><mo>)</mo></mrow></mtd></mtr></mtable></math></maths><img file="US10791355B2_D0006.tif" />
0064The expected shares E[S<sub>i</sub>] computed by the example shares calculator <b>212</b> represents the condition probability that given that the panelist is watching television, then the panelist and, thus, the persons the panelist represents, is watching the ith program (e.g., first, second, or third programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c</i>).
0065Referring to Table 2 above including the probabilities of viewership activity with respect to the first, second, and/or third programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c</i>, the example shares calculator <b>212</b> calculates the expected shares <b>213</b> for the first program <b>112</b><i>a</i>, the second program <b>112</b><i>b</i>, and the third program <b>112</b><i>c </i>using Equations (11) or (14). For example, the shares calculator <b>212</b> can calculate the following expected shares <b>213</b> for first, second, and/or third programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c </i>as follows: <br /><i>E[S</i><sub>i</sub>]=[0.3404 0.3907 0.2689] (15)
0066Also, the example shares calculator <b>212</b> can calculate a covariance matrix σ<sup>2</sup>(S<sub>i</sub>, S<sub>j</sub>) based on the variance (e.g., Equation (12)) and the covariance (e.g., Equation (13)) for Table 2 as follows:
0067<maths id="MATH-US-00007" num="00007"><math overflow="scroll"><mtable><mtr><mtd><mrow><mrow><msup><mi>σ</mi><mn>2</mn></msup><mo></mo><mrow><mo>(</mo><mrow><msub><mi>S</mi><mi>i</mi></msub><mo></mo><msub><mi>S</mi><mi>j</mi></msub></mrow><mo>)</mo></mrow></mrow><mo>=</mo><mrow><mo>(</mo><mtable><mtr><mtd><mn>0.0293</mn></mtd><mtd><mrow><mo>-</mo><mn>0.0146</mn></mrow></mtd><mtd><mrow><mo>-</mo><mn>0.0147</mn></mrow></mtd></mtr><mtr><mtd><mrow><mo>-</mo><mn>0.0146</mn></mrow></mtd><mtd><mn>0.0299</mn></mtd><mtd><mrow><mo>-</mo><mn>0.0153</mn></mrow></mtd></mtr><mtr><mtd><mrow><mo>-</mo><mn>0.0147</mn></mrow></mtd><mtd><mrow><mo>-</mo><mn>0.0153</mn></mrow></mtd><mtd><mn>0.0300</mn></mtd></mtr></mtable><mo>)</mo></mrow></mrow></mtd><mtd><mrow><mo>(</mo><mn>16</mn><mo>)</mo></mrow></mtd></mtr></mtable></math></maths><img file="US10791355B2_D0007.tif" />
0068The covariance matrix (16) indicates relationships between, for example, the first program <b>112</b><i>a </i>and the other programs <b>112</b><i>b</i>, <b>112</b><i>c</i>. In the example covariance matrix (16), the diagonals of the matrix (16) are computed by the shares calculator <b>212</b> based on the variance (e.g., Equation (12)) and the off-diagonals of the matrix are computed based on the covariance (e.g., Equation (13)). In the example covariance matrix (16), the off-diagonals include negative values. The negative values of the off-diagonals in the covariance matrix (16) reflect the fact out of the population who is viewing television, more people in the population who are watching one program (e.g., the first program <b>112</b><i>a</i>) means that less people in the population are watching the other programs (e.g., the second program <b>112</b><i>b</i>, the third program <b>112</b><i>c</i>). Thus, the example shares calculator <b>212</b> of <figref idref="DRAWINGS">FIG. 2</figref> calculates the shares for the first, second, and third programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c </i>based the probabilities that the panelists (and, thus, the population the panelists represent) are viewing the television and viewing certain programs.
0069Thus, the ratings calculator <b>210</b> and the shares calculator <b>212</b> of the example viewing activity analyzer <b>132</b> of <figref idref="DRAWINGS">FIG. 2</figref> determines ratings and/or shares for one or more of the programs <b>112</b><i>a</i>-<b>112</b><i>n </i>that may be viewed by panelist, such as the first panelist <b>104</b> and/or the second panelist <b>118</b> of <figref idref="DRAWINGS">FIG. 1</figref>. As disclosed above, each of the panelists <b>104</b>, <b>118</b> is associated with respective demographics <b>114</b>, <b>124</b>. The example viewing activity analyzer <b>132</b> of <figref idref="DRAWINGS">FIG. 2</figref> can also calculate the ratings and/or shares for the program(s) <b>112</b><i>a</i>-<b>112</b><i>n </i>based on a subgroup of interest, such as a subgroup associated with a particular demographic (e.g., age, gender).
0070The example viewing activity analyzer of <figref idref="DRAWINGS">FIG. 2</figref> includes a subgroup analyzer <b>214</b>. In some examples, a user of the example processor <b>126</b> of <figref idref="DRAWINGS">FIG. 1</figref> can request that the ratings calculator <b>210</b> of the viewing activity analyzer <b>132</b> calculate ratings <b>211</b> for one or more of the programs <b>112</b><i>a</i>-<b>112</b><i>n </i>for a particular demographic group (e.g., by providing a user input to the processor <b>126</b>). Additionally or alternatively, the user can request that the shares calculator <b>212</b> calculate shares <b>213</b> for one or more of the programs <b>112</b><i>a</i>-<b>112</b><i>n </i>for a particular demographic group. The example subgroup analyzer <b>214</b> identifies the relevant demographics <b>114</b>, <b>124</b> of the data streams <b>128</b>, <b>130</b> stored in the example database <b>202</b> of the viewing activity analyzer <b>132</b>. The subgroup analyzer <b>214</b> provides the relevant demographic data to the ratings calculator <b>210</b> and/or the shares calculator <b>212</b>.
0071For example, referring to Table 1 above, a user may be interested in ratings <b>211</b> and shares <b>213</b> for one or more of the programs <b>112</b><i>a</i>-<b>112</b><i>n </i>for just the “young” demographic group. Based on a user input received by the processor <b>126</b> directing the viewing activity analyzer <b>132</b> to determine the ratings for the “young” demographic group, the example subgroup analyzer <b>214</b> identifies the relevant data streams <b>128</b>, <b>130</b> stored in the database <b>202</b> corresponding to the demographic group of interest. For example, with respect to the “young” demographic group, the subgroup analyzer <b>214</b> identifies the viewing data associated with Panelist A (e.g., the first panelist <b>104</b>), Panelist B (e.g., the second panelist <b>118</b>), and Panelist C based on their association with the demographic group of interest. In some examples, the subgroup analyzer <b>214</b> scans the data stored in the database <b>202</b> to identify the relevant panelist viewing data based on, for example, tags associated with the data stream <b>128</b>, <b>130</b> stored in the database <b>202</b>.
0072The example subgroup analyzer <b>214</b> provides the relevant viewing data for the demographic group of interest to the ratings calculator <b>210</b> and the shares calculator <b>212</b>. The example ratings calculator <b>210</b> of <figref idref="DRAWINGS">FIG. 2</figref> applies one or more of Equations (1)-(6) above to determine the expected rating(s) <b>211</b> for the programs(s) <b>112</b><i>a</i>-<b>112</b><i>n </i>for the selected demographic group. For example, the ratings calculator <b>210</b> performs the summations for only the demographic group of interest (e.g., E[R<sub>i</sub>]=Σ<sub>k=1</sub>v<sub>k</sub>p<sub>k,i</sub>, where n is the number of panelists in the “young” demographic group). In some examples, the ratings calculator <b>210</b> determines normalized weights (e.g., the weight v<sub>k</sub>) for the demographic group of interest based on the sampling weights <b>205</b> assigned to the panelists associated with the demographic group of interest. The rating calculator <b>210</b> uses the normalized weights to calculate the expected ratings, variance, and/or covariance for the demographic group of interest.
0073Similarly, the example shares calculator <b>212</b> of <figref idref="DRAWINGS">FIG. 2</figref> calculates the share(s) <b>213</b> for the program(s) <b>112</b><i>a</i>-<b>112</b><i>n </i>using the normalized weights for the demographic group of interest and by summing across the number of panelists in the demographic group of interest to calculate the expected share(s) <b>213</b>, variance, and/or covariance (e.g., using Equations (9)-(14)). Thus, the example viewing activity analyzer <b>132</b> can determine expected ratings and/or shares and respective variance and covariance of the ratings and/or shares for a subgroup of interest.
0074The example subgroup analyzer <b>214</b> can also determine one or more subgroup viewing metrics <b>217</b>. For example, the subgroup analyzer <b>214</b> can determine a probability that a person within a demographic group of interest is watching a particular program <b>112</b><i>a</i>-<b>112</b><i>n </i>(e.g. in response to user input received by the processor <b>126</b>). For example, a user may be interested in a probability that a person in the “middle” demographic age group of Table 1 is watching the one of the programs <b>112</b><i>a</i>, <b>112</b><i>b</i>, <b>112</b><i>c</i>. In such examples, the example subgroup analyzer <b>214</b> of <figref idref="DRAWINGS">FIG. 2</figref> identifies Panelists D, E, and F as associated with the “middle” age group and generates a vector K including viewing data for Panelists D, E, and F from the respective data streams stored in the database <b>202</b>. For example, based on the data in the vector K, the subgroup analyzer <b>214</b> can determine the probability that any person within the “middle” age group is watching i<sup>th </sup>program as follows:
0075Prob[K∈ith program]=1−Π<sub>k∈K</sub>(1−p<sub>k,i</sub>) (17), where Prob[K∈ith program] is the probabitlity at least one person in the subgroup of interest is watching the i<sup>th </sup>program and (1−p<sub>k,i</sub>) is the probability a person in the subgroup is not watching the i<sup>th </sup>program.
0076Thus, Equation (17) calculates a product over members of the selected subgroup with respect to the subgroup members watching a program of interest.
0077For example, referring to Table 1, the probability identifier <b>208</b> identified a 20% probability that Panelist D is watching the first program <b>112</b><i>a</i>, a 0% probability that Panelist E is watching the first program <b>112</b><i>a</i>, and a 30% probability that Panelist F is watching the first program <b>112</b><i>a</i>. The subgroup analyzer <b>214</b> can determine the probability that one of Panelists D, E, or F are watching the first program <b>112</b><i>a </i>as follows: <br />Prob[“Middle” age group watching first program]=1−(1−0.2)(1−0)(1−0.3)=0.44 (18)
0078Thus, the subgroup analyzer <b>214</b> determines that there is a 44% probability that a panelist (and, thus, the persons the panelist(s) represent) in the “middle” age demographic is watching the first program <b>112</b><i>a</i>. Also, subgroup analyzer <b>214</b> can determine the variance as follows: <br />Var[<i>X</i>]=(Π<sub>k∈K</sub>(1−<i>p</i><sub>k,i</sub>))(1−Π<sub>k∈K</sub>(1−<i>p</i><sub>k,i</sub>)) (19)
0079The example subgroup analyzer <b>214</b> of <figref idref="DRAWINGS">FIG. 2</figref> can also determine for a given program <b>112</b><i>a</i>-<b>112</b><i>n</i>, a percentage of people watching the program who are associated with a certain demographic. In the example of <figref idref="DRAWINGS">FIG. 2</figref>, data regarding the number of panelists in a demographic who are viewing the program <b>112</b><i>a</i>-<b>112</b><i>n </i>of interest and the number of total people viewing the program <b>112</b><i>a</i>-<b>112</b><i>n </i>of interest are random variables. The subgroup analyzer <b>214</b> analyzes different probabilistic combinations of groups of panelists (e.g., the panelists in Table 1) who are watching the program <b>112</b><i>a</i>-<b>112</b><i>n </i>of interest and the respective sampling weights <b>205</b> assigned to the panelists. In some examples, the subgroup analyzer <b>214</b> approximates the percentage of panelists in a demographic group of interest who are watching the program <b>112</b><i>a</i>-<b>112</b><i>n </i>of interest based on a large panel size (e.g., thousands of panelists). For example, for a group of K people, the subgroup analyzer <b>214</b> can approximate a proportion of panelists watching program i that belong to the group K as follows:
0080<maths id="MATH-US-00008" num="00008"><math overflow="scroll"><mtable><mtr><mtd><mrow><msub><mi>p</mi><mrow><mo>{</mo><mi>K</mi><mo>}</mo></mrow></msub><mo>=</mo><mfrac><mrow><munder><mo>∑</mo><mrow><mi>k</mi><mo>∈</mo><mi>K</mi></mrow></munder><mo></mo><mrow><msub><mi>w</mi><mi>k</mi></msub><mo></mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>i</mi></mrow></msub></mrow></mrow><mrow><munderover><mo>∑</mo><mrow><mi>k</mi><mo>=</mo><mn>1</mn></mrow><mi>n</mi></munderover><mo></mo><mrow><msub><mi>w</mi><mi>k</mi></msub><mo></mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>i</mi></mrow></msub></mrow></mrow></mfrac></mrow></mtd><mtd><mrow><mo>(</mo><mn>20</mn><mo>)</mo></mrow></mtd></mtr></mtable></math></maths><img file="US10791355B2_D0008.tif" />
0081In Equation (20), above, the numerator represents the subgroup of interest and the denominator considers all panelists viewing the program of interest (e.g., all demographics). The subgroup analyzer <b>214</b> can determine the variance and covariance as follows:
0082<maths id="MATH-US-00009" num="00009"><math overflow="scroll"><mtable><mtr><mtd><mrow><mrow><mi>Var</mi><mo></mo><mrow><mo>[</mo><msub><mi>p</mi><mrow><mrow><mo>{</mo><mi>K</mi><mo>}</mo></mrow><mo>,</mo><mi>i</mi></mrow></msub><mo>]</mo></mrow></mrow><mo>=</mo><mfrac><mrow><munder><mo>∑</mo><mrow><mi>k</mi><mo>∈</mo><mi>K</mi></mrow></munder><mo></mo><mrow><mrow><msubsup><mi>w</mi><mi>k</mi><mn>2</mn></msubsup><mo></mo><mrow><mo>(</mo><mrow><mn>1</mn><mo>-</mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>i</mi></mrow></msub></mrow><mo>)</mo></mrow></mrow><mo></mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>i</mi></mrow></msub></mrow></mrow><msup><mrow><mo>(</mo><mrow><munderover><mo>∑</mo><mrow><mi>k</mi><mo>=</mo><mn>1</mn></mrow><mi>n</mi></munderover><mo></mo><mrow><msub><mi>w</mi><mi>k</mi></msub><mo></mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>i</mi></mrow></msub></mrow></mrow><mo>)</mo></mrow><mn>2</mn></msup></mfrac></mrow></mtd><mtd><mrow><mo>(</mo><mn>21</mn><mo>)</mo></mrow></mtd></mtr><mtr><mtd><mrow><mrow><mi>Cov</mi><mo></mo><mrow><mo>[</mo><mrow><msub><mi>p</mi><mrow><mrow><mo>{</mo><mi>K</mi><mo>}</mo></mrow><mo>,</mo><mi>i</mi></mrow></msub><mo>,</mo><msub><mi>p</mi><mrow><mrow><mo>{</mo><mi>K</mi><mo>}</mo></mrow><mo>,</mo><mi>j</mi></mrow></msub></mrow><mo>]</mo></mrow></mrow><mo>=</mo><mrow><mo>-</mo><mfrac><mrow><mrow><munder><mo>∑</mo><mrow><mi>k</mi><mo>∈</mo><mi>K</mi></mrow></munder><mo></mo><mrow><msubsup><mi>w</mi><mi>k</mi><mn>2</mn></msubsup><mo></mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>i</mi></mrow></msub></mrow></mrow><mo>,</mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>j</mi></mrow></msub></mrow><mrow><mrow><mo>(</mo><mrow><munderover><mo>∑</mo><mrow><mi>k</mi><mo>=</mo><mn>1</mn></mrow><mi>n</mi></munderover><mo></mo><mrow><msub><mi>w</mi><mi>k</mi></msub><mo></mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>i</mi></mrow></msub></mrow></mrow><mo>)</mo></mrow><mo></mo><mrow><mo>(</mo><mrow><munderover><mo>∑</mo><mrow><mi>k</mi><mo>=</mo><mn>1</mn></mrow><mi>n</mi></munderover><mo></mo><mrow><msub><mi>w</mi><mi>k</mi></msub><mo></mo><msub><mi>p</mi><mrow><mi>k</mi><mo>,</mo><mi>j</mi></mrow></msub></mrow></mrow><mo>)</mo></mrow></mrow></mfrac></mrow></mrow></mtd><mtd><mrow><mo>(</mo><mn>22</mn><mo>)</mo></mrow></mtd></mtr></mtable></math></maths><img file="US10791355B2_D0009.tif" />
0083The covariance determined by Equation (22) can be used to analyze viewing activity for different programs <b>112</b><i>a</i>-<b>112</b><i>n </i>across the subgroup of interest. For example, the covariance can be analyzed with respect to a proportion of viewers belonging to a subgroup across two different programs <b>112</b><i>a</i>-<b>112</b><i>n. </i>
0084The example viewing activity analyzer <b>132</b> can calculate the ratings <b>211</b>, the shares <b>213</b>, and/or the subgroup viewing metrics <b>217</b> at the household level in addition or as an alternative to determining viewing metrics at the panelist level or demographic group level. For example, the sampling weight assigner <b>204</b> can assign sampling weights <b>205</b> to the first household <b>102</b> and/or the second household <b>116</b> based on, for example, household size. Based on a user request to calculate, for example, ratings <b>211</b> and/or shares <b>213</b> at the household level, the subgroup analyzer <b>214</b> can identify and/or format the viewing data of the data streams <b>128</b>, <b>130</b> by household. As an example, the ratings calculator <b>210</b> can determine the ratings <b>211</b> based on a probability that any member of the household (e.g., the first household <b>102</b>) is watching television.
0085Thus, the example viewing activity analyzer <b>132</b> can determine different viewing activity metrics such as ratings <b>211</b> and/or shares <b>213</b> despite probabilities or uncertainties in the data streams (e.g., the data streams <b>128</b>, <b>130</b> of <figref idref="DRAWINGS">FIG. 1</figref>) received from the panel meters (e.g., the meters <b>108</b>, <b>122</b>). The example viewing activity analyzer <b>132</b> can also determine subgroup-specific metrics, including, for example, what program(s) a demographic group is watching and/or what demographic group is watching a certain program. The example viewing activity analyzer <b>132</b> includes a communicator <b>216</b>. The communicator <b>216</b> outputs one or more of the ratings <b>211</b>, shares <b>213</b>, or subgroup metrics (e.g., the viewing metric output(s) <b>134</b> of <figref idref="DRAWINGS">FIG. 1</figref>) for display via, for example, the output device <b>136</b>.
0086While an example manner of implementing the viewing activity analyzer <b>132</b> is illustrated in <figref idref="DRAWINGS">FIGS. 1-2</figref>, one or more of the elements, processes and/or devices illustrated in <figref idref="DRAWINGS">FIGS. 1-2</figref> may be combined, divided, re-arranged, omitted, eliminated and/or implemented in any other way. Further, the example data collector <b>200</b>, the example database <b>202</b>, the example sampling weight assigner <b>204</b>, the example probability identifier <b>208</b>, the example ratings calculator <b>210</b>, the example shares calculator <b>212</b>, the example subgroup analyzer <b>214</b>, the example communicator <b>216</b> and/or, more generally, the example viewing activity analyzer of <figref idref="DRAWINGS">FIGS. 1-2</figref> may be implemented by hardware, software, firmware and/or any combination of hardware, software and/or firmware. Thus, for example, any of the example data collector <b>200</b>, the example database <b>202</b>, the example sampling weight assigner <b>204</b>, the example probability identifier <b>208</b>, the example ratings calculator <b>210</b>, the example shares calculator <b>212</b>, the example subgroup analyzer <b>214</b>, the example communicator <b>216</b> and/or, more generally, the example viewing activity analyzer of <figref idref="DRAWINGS">FIGS. 1-2</figref> could be implemented by one or more analog or digital circuit(s), logic circuits, programmable processor(s), application specific integrated circuit(s) (ASIC(s)), programmable logic device(s) (PLD(s)) and/or field programmable logic device(s) (FPLD(s)). When reading any of the apparatus or system claims of this patent to cover a purely software and/or firmware implementation, at least one of the example data collector <b>200</b>, the example database <b>202</b>, the example sampling weight assigner <b>204</b>, the example probability identifier <b>208</b>, the example ratings calculator <b>210</b>, the example shares calculator <b>212</b>, the example subgroup analyzer <b>214</b>, the example communicator <b>216</b> and/or, more generally, the example viewing activity analyzer of <figref idref="DRAWINGS">FIGS. 1-2</figref> is/are hereby expressly defined to include a tangible computer readable storage device or storage disk such as a memory, a digital versatile disk (DVD), a compact disk (CD), a Blu-ray disk, etc. storing the software and/or firmware. Further still, the example data collector <b>200</b>, the example database <b>202</b>, the example sampling weight assigner <b>204</b>, the example probability identifier <b>208</b>, the example ratings calculator <b>210</b>, the example shares calculator <b>212</b>, the example subgroup analyzer <b>214</b>, the example communicator <b>216</b> and/or, more generally, the example viewing activity analyzer of <figref idref="DRAWINGS">FIGS. 1-2</figref> may include one or more elements, processes and/or devices in addition to, or instead of, those illustrated in <figref idref="DRAWINGS">FIGS. 1-2</figref>, and/or may include more than one of any or all of the illustrated elements, processes and devices.
0087A flowchart representative of example machine readable instructions for implementing the example viewing activity analyzer <b>132</b> of <figref idref="DRAWINGS">FIGS. 1-2</figref> is shown in <figref idref="DRAWINGS">FIG. 3</figref>. In this example, the machine readable instructions comprise a program for execution by a processor such as the processor <b>126</b> of <figref idref="DRAWINGS">FIG. 1</figref> and shown in the example processor platform <b>400</b> discussed below in connection with <figref idref="DRAWINGS">FIG. 4</figref>. The program may be embodied in software stored on a tangible computer readable storage medium such as a CD-ROM, a floppy disk, a hard drive, a digital versatile disk (DVD), a Blu-ray disk, or a memory associated with the processor <b>126</b>, but the entire program and/or parts thereof could alternatively be executed by a device other than the processor <b>126</b> and/or embodied in firmware or dedicated hardware. Further, although the example program is described with reference to the flowchart illustrated in <figref idref="DRAWINGS">FIG. 3</figref>, many other methods of implementing the example viewing activity analyzer <b>132</b> of <figref idref="DRAWINGS">FIGS. 1-2</figref> may alternatively be used. For example, the order of execution of the blocks may be changed, and/or some of the blocks described may be changed, eliminated, or combined.
0088As mentioned above, the example process of <figref idref="DRAWINGS">FIG. 3</figref> may be implemented using coded instructions (e.g., computer and/or machine readable instructions) stored on a tangible computer readable storage medium such as a hard disk drive, a flash memory, a read-only memory (ROM), a compact disk (CD), a digital versatile disk (DVD), a cache, a random-access memory (RAM) and/or any other storage device or storage disk in which information is stored for any duration (e.g., for extended time periods, permanently, for brief instances, for temporarily buffering, and/or for caching of the information). As used herein, the term “tangible computer readable storage medium” is expressly defined to include any type of computer readable storage device and/or storage disk and to exclude propagating signals and to exclude transmission media. As used herein, “tangible computer readable storage medium” and “tangible machine readable storage medium” are used interchangeably. Additionally or alternatively, the example process of <figref idref="DRAWINGS">FIG. 3</figref> may be implemented using coded instructions (e.g., computer and/or machine readable instructions) stored on a non-transitory computer and/or machine readable medium such as a hard disk drive, a flash memory, a read-only memory, a compact disk, a digital versatile disk, a cache, a random-access memory and/or any other storage device or storage disk in which information is stored for any duration (e.g., for extended time periods, permanently, for brief instances, for temporarily buffering, and/or for caching of the information). As used herein, the term non-transitory computer readable medium is expressly defined to include any type of computer readable storage device and/or storage disk and to exclude propagating signals and to exclude transmission media. As used herein, when the phrase “at least” is used as the transition term in a preamble of a claim, it is open-ended in the same manner as the term “comprising” is open ended.
0089The program <b>300</b> of <figref idref="DRAWINGS">FIG. 3</figref> begins at block <b>302</b> with the data collector <b>200</b> of the example viewing activity analyzer <b>132</b> of <figref idref="DRAWINGS">FIG. 2</figref> accessing one or more data streams such as the first data stream <b>128</b> and/or the second data stream <b>130</b> of <figref idref="DRAWINGS">FIG. 1</figref> from the panel meter(s) <b>108</b>, <b>122</b> associated with televisions <b>106</b>, <b>120</b> in one or more households <b>102</b>, <b>116</b> (block <b>302</b>). The data stream(s) <b>128</b>, <b>132</b> include television viewing data with respect to, for example, a program <b>112</b><i>a</i>-<b>112</b><i>n </i>broadcast by the television(s) <b>106</b>, <b>120</b> and viewed by the panelists <b>104</b>, <b>118</b>. The data can be stored in the example database <b>202</b> of the example viewing activity analyzer of <figref idref="DRAWINGS">FIG. 2</figref>.
0090The program of <figref idref="DRAWINGS">FIG. 3</figref> includes the example sampling weight assigner <b>204</b> of <figref idref="DRAWINGS">FIG. 2</figref> assigning sampling weight(s) <b>205</b> to the panelist(s) <b>104</b>, <b>118</b> (block <b>304</b>). The sampling weight assigner <b>204</b> can assign the sampling weight(s) <b>205</b> based on, for example, one or more demographics <b>114</b>, <b>124</b> associated with the panelist(s) <b>104</b>, <b>118</b>, such as age and/or gender. In some examples, the sampling weight assigner <b>204</b> assigns sampling weight(s) <b>205</b> to the household(s) <b>102</b>, <b>116</b> from which the data stream(s) <b>218</b>, <b>130</b> are received based on, for example, household size. The sampling weight assigner <b>204</b> can assign the sampling weight(s) <b>205</b> based on one or more sampling weight rule(s) <b>206</b> stored in the example database <b>202</b> of the viewing activity analyzer <b>132</b> of <figref idref="DRAWINGS">FIG. 2</figref>.
0091The program of <figref idref="DRAWINGS">FIG. 3</figref> includes the probability identifier <b>208</b> determining viewing probabilities <b>209</b> for, for example, the panelist(s) <b>104</b>, <b>116</b> associated with data stream(s) <b>128</b>, <b>130</b> (block <b>306</b>). The probability identifier <b>208</b> identifies uncertainties in the data stream(s) <b>128</b>, <b>130</b> with respect to, for example, whether a panelist is watching television, what program <b>112</b><i>a</i>-<b>112</b><i>n </i>the panelist is watching, etc. In some examples, the uncertainties are due to, for example, potential co-viewing activity between two or more members of a household. In other examples, the uncertainties are due to, for example, a technical error in the collection of the viewing data by the panel meter(s) <b>108</b>, <b>122</b>. The probability identifier <b>208</b> can determine the viewing probabilities <b>209</b> with respect to whether or not a panelist watched television and/or what program(s) <b>112</b><i>a</i>-<b>112</b><i>n </i>the panelist could have watched based on one or more probability rule(s) <b>207</b> stored in the example database <b>202</b> of <figref idref="DRAWINGS">FIG. 2</figref>. In some examples, the probability identifier <b>208</b> identifies the probabilities <b>209</b> with respect to, for example, whether or not any member of a household is watching television.
0092The program of <figref idref="DRAWINGS">FIG. 3</figref> includes the example ratings calculator <b>210</b> of <figref idref="DRAWINGS">FIG. 2</figref> calculating expected ratings <b>211</b> for one or more programs <b>112</b><i>a</i>-<b>112</b><i>n </i>(block <b>308</b>). In some examples, the ratings calculator <b>210</b> uses one or more algorithms to calculate the ratings <b>211</b>, such as Equations (1) or (4) disclosed above. In determining the ratings <b>211</b>, the ratings calculator <b>210</b> accounts for the probabilities <b>209</b> with respect to whether or not the panelist(s) <b>104</b>, <b>118</b> are watching television, what program(s) <b>112</b><i>a</i>-<b>112</b><i>n </i>the panelist(s) <b>104</b>, <b>118</b> are watching, etc. In some examples, the ratings calculator <b>210</b> calculates the variance (e.g., using Equations (2), (5)) and/or covariance (e.g., using Equations (3), (6)) with respect to program viewing activity to analyze viewership behavior between, for example, two or more programs <b>112</b><i>a</i>-<b>112</b><i>n </i>(e.g., as reflected in the example covariance matrix (8)). In some examples, the ratings calculator <b>210</b> calculates a null rating <b>211</b> indicative of a percentage of panelists who are not watching any program.
0093In some examples of the program of <figref idref="DRAWINGS">FIG. 3</figref>, the shares calculator <b>212</b> of <figref idref="DRAWINGS">FIG. 2</figref> additionally or alternatively calculates the expected share(s) <b>213</b> for the program(s) <b>112</b><i>a</i>-<b>112</b><i>n </i>(e.g., using Equation (11) disclosed above) (block <b>308</b>). For example, to calculate the share(s) <b>213</b>, the example shares calculator <b>212</b> adjusts the sampling weight(s) <b>205</b> assigned to the panelist(s) <b>104</b>, <b>118</b> to determine share weight(s) <b>215</b> representative of television viewing behavior by the panelist(s) <b>104</b>, <b>118</b> (e.g., based on probabilities <b>209</b> indicating that the panelist(s) <b>104</b>, <b>118</b> may or may not be watching television). The example shares calculator <b>212</b> calculates the shares <b>213</b> based on the share weights <b>215</b> and conditional probabilities that the panelist(s) are watching a particular program <b>112</b><i>a</i>-<b>112</b><i>n</i>, if the panelist(s) are watching television (e.g., determined based on the probabilities <b>209</b> with respect to program viewing probabilities). In some examples, the shares calculator <b>212</b> calculates the variance (e.g., using Equation (12)) and/or covariance (e.g., using Equation (13)) with respect to program viewing activity to analyze viewing activity between, for example, two or more programs <b>112</b><i>a</i>-<b>112</b><i>n </i>(e.g., as reflected in the example covariance matrix (16)).
0094The example of <figref idref="DRAWINGS">FIG. 3</figref> includes a determination as to whether the example subgroup analyzer <b>214</b> of <figref idref="DRAWINGS">FIG. 2</figref> is to calculate one or more subgroup viewing metrics <b>217</b> (block <b>310</b>). In some examples, the subgroup analyzer <b>214</b> calculates the subgroup viewing metric(s) <b>217</b> based on one or more user inputs received via the processor <b>126</b> of <figref idref="DRAWINGS">FIG. 1</figref> that instructs viewing metrics such as ratings <b>211</b> and/or shares <b>213</b> to be calculated for one or more demographic groups of interest (e.g., an age group, an ethnic group, a gender group).
0095The subgroup analyzer <b>214</b> identifies viewing data for the subgroup of interest based on the data streams <b>128</b>, <b>130</b> stored in the database <b>202</b> of <figref idref="DRAWINGS">FIG. 2</figref> (block <b>312</b>). In some examples, the subgroup analyzer <b>214</b> identifies the relevant viewing data for panelists in the subgroup of interest (e.g., the panelists <b>104</b>, <b>118</b> of <figref idref="DRAWINGS">FIG. 1</figref>) based on one or more tags identify the demographics <b>114</b>, <b>124</b> associated with the panelists.
0096The example of <figref idref="DRAWINGS">FIG. 3</figref> includes calculating the one or more subgroup viewing metrics <b>217</b> (block <b>314</b>). In some examples, the subgroup analyzer <b>214</b> instructs the ratings calculator <b>210</b> to calculate ratings <b>211</b> for one or more programs <b>112</b><i>a</i>-<b>112</b><i>n </i>for the subgroup of interest. In some examples, the subgroup analyzer <b>214</b> instructs the shares calculator <b>212</b> to calculate shares <b>213</b> for one or more programs <b>112</b><i>a</i>-<b>112</b><i>n </i>for the subgroup of interest. In such examples, the ratings calculator <b>210</b> calculates the ratings <b>211</b> (and, in some examples, the variance and covariance) substantially as disclosed above (e.g., at block <b>308</b>) for the subgroup of interest. Also in some such examples, the shares calculator <b>212</b> calculates the shares <b>213</b> and, in some examples, the variance and covariance) substantially as disclosed above (e.g., at block <b>308</b>) for the subgroup of interest.
0097In some examples, the subgroup analyzer <b>214</b> calculates subgroup viewing metrics <b>217</b> with respect to, for example, a probability that a subgroup of interest is watching one or more of the programs <b>112</b><i>a</i>-<b>112</b><i>n</i>. For example, the subgroup analyzer <b>214</b> uses Equation (17), disclosed above, to determine a probability that any person within a demographic group of interest is watching one of the programs <b>112</b><i>a</i>-<b>112</b><i>n</i>. In some examples, the subgroup analyzer <b>214</b> determines a subgroup that is watching a particular program <b>112</b><i>a</i>-<b>112</b><i>n</i>. For example, the subgroup analyzer <b>214</b> uses Equation (20), disclosed above, to approximate a proportion of panelists watching one of the programs <b>112</b><i>a</i>-<b>112</b><i>n </i>that belong to a subgroup of interest (e.g., a demographic group of interest).
0098If a decision is made not to calculate viewing metrics for a subgroup (e.g., at block <b>310</b>), the example program <b>300</b> ends. Also, if there are no further subgroup viewing metrics <b>217</b> to calculate (e.g., based on user input(s) received at the processor <b>126</b>), the example program <b>300</b> ends.
0099<figref idref="DRAWINGS">FIG. 4</figref> is a block diagram of an example processor platform <b>400</b> capable of executing the instructions of <figref idref="DRAWINGS">FIG. 3</figref> to implement the data collector <b>200</b>, the example database <b>202</b>, the example sampling weight assigner <b>204</b>, the example probability identifier <b>208</b>, the example ratings calculator <b>210</b>, the example shares calculator <b>212</b>, the example subgroup analyzer <b>214</b>, the example communicator <b>216</b> and/or, more generally, the example viewing activity analyzer of <figref idref="DRAWINGS">FIGS. 1-2</figref>. The processor platform <b>400</b> can be, for example, a server, a personal computer, a mobile device (e.g., a cell phone, a smart phone, a tablet such as an iPad™), a personal digital assistant (PDA), an Internet appliance, a set top box, or any other type of computing device.
0100The processor platform <b>400</b> of the illustrated example includes the processor <b>126</b>. The processor <b>126</b> of the illustrated example is hardware. For example, the processor <b>126</b> can be implemented by one or more integrated circuits, logic circuits, microprocessors or controllers from any desired family or manufacturer.
0101The processor <b>126</b> of the illustrated example includes a local memory <b>413</b> (e.g., a cache). The processor <b>126</b> of the illustrated example is in communication with a main memory including a volatile memory <b>414</b> and a non-volatile memory <b>416</b> via a bus <b>418</b>. The volatile memory <b>414</b> may be implemented by Synchronous Dynamic Random Access Memory (SDRAM), Dynamic Random Access Memory (DRAM), RAMBUS Dynamic Random Access Memory (RDRAM) and/or any other type of random access memory device. The non-volatile memory <b>416</b> may be implemented by flash memory and/or any other desired type of memory device. Access to the main memory <b>414</b>, <b>416</b> is controlled by a memory controller.
0102The processor platform <b>400</b> of the illustrated example also includes an interface circuit <b>420</b>. The interface circuit <b>420</b> may be implemented by any type of interface standard, such as an Ethernet interface, a universal serial bus (USB), and/or a PCI express interface.
0103In the illustrated example, one or more input devices <b>422</b> are connected to the interface circuit <b>420</b>. The input device(s) <b>422</b> permit(s) a user to enter data and commands into the processor <b>126</b>. The input device(s) can be implemented by, for example, an audio sensor, a microphone, a camera (still or video), a keyboard, a button, a mouse, a touchscreen, a track-pad, a trackball, isopoint and/or a voice recognition system.
0104One or more output devices <b>136</b>, <b>424</b> are also connected to the interface circuit <b>420</b> of the illustrated example. The output devices <b>136</b>, <b>424</b> can be implemented, for example, by display devices (e.g., a light emitting diode (LED), an organic light emitting diode (OLED), a liquid crystal display, a cathode ray tube display (CRT), a touchscreen, a tactile output device, a printer and/or speakers). The interface circuit <b>420</b> of the illustrated example, thus, typically includes a graphics driver card, a graphics driver chip or a graphics driver processor.
0105The interface circuit <b>420</b> of the illustrated example also includes a communication device such as a transmitter, a receiver, a transceiver, a modem and/or network interface card to facilitate exchange of data with external machines (e.g., computing devices of any kind) via a network <b>426</b> (e.g., an Ethernet connection, a digital subscriber line (DSL), a telephone line, coaxial cable, a cellular telephone system, etc.).
0106The processor platform <b>400</b> of the illustrated example also includes one or more mass storage devices <b>428</b> for storing software and/or data. Examples of such mass storage devices <b>428</b> include floppy disk drives, hard drive disks, compact disk drives, Blu-ray disk drives, RAID systems, and digital versatile disk (DVD) drives.
0107Coded instructions <b>432</b> to implement the instructions of <figref idref="DRAWINGS">FIG. 3</figref> may be stored in the mass storage device <b>428</b>, in the volatile memory <b>1014</b>, in the non-volatile memory <b>416</b>, and/or on a removable tangible computer readable storage medium such as a CD or DVD.
0108From the foregoing, it will be appreciated that the above disclosed systems, methods, and apparatus improves the ability to determine viewing metrics such as ratings and/or shares for media such as one or more television programs in view of uncertainties or probabilities in the data from which the viewing metrics are calculated. Examples disclosed herein determines the viewing metrics by accounting for different scenarios with respect to whether a panelist is watching television, what program he or she is watching, etc. and the probabilities that such scenarios will happen. Examples disclosed herein compute expected ratings and/or expected shares and respective variance or covariance thereof despite the probabilities in the viewing data. Thus, examples disclosed herein compute ratings and/or shares that more accurately reflect viewer behavior as compared to ratings and/or shares calculated based on the randomly assigned probability data (e.g., the 0's and 1's).
0109Examples disclosed herein increase efficiency and reduce processor resources in determining the ratings and/or shares based on the probabilistic data as compared to, for example, repeating probabilistic stimulations thousands of times, by approximating expected ratings and/or shares. Some disclosed examples provide for calculation of subgroup-specific metrics. Disclosed examples provide accurate and efficient analyses of viewing behavior despite uncertainties or probabilities in the viewing data.
0110Although certain example methods, apparatus and articles of manufacture have been disclosed herein, the scope of coverage of this patent is not limited thereto. On the contrary, this patent covers all methods, apparatus and articles of manufacture fairly falling within the scope of the claims of this patent.
Contents4
25 sheets
Sheet 1 Sheet 2 Sheet 3 Sheet 4 Sheet 5 Sheet 6 Sheet 7 Sheet 8 Sheet 9 Sheet 10 Sheet 11 Sheet 12 Sheet 13 Sheet 14 Sheet 15 Sheet 16 Sheet 17 Sheet 18 Sheet 19 Sheet 20 Sheet 21 Sheet 22 Sheet 23 Sheet 24 Sheet 25
Every citation, both ways
| Document | Relation | Office | Cited during |
|---|---|---|---|
| US11516543B2 | Cited by | United States of America | Applicant |
| US11778255B2 | Cited by | United States of America | Search report |
| US2025168430A1 | Cited by | United States of America | Search report |
| US11785301B2 | Cited by | United States of America | Applicant |
| WO0111606A1 | Cites | World Intellectual Property Organization (WIPO) | Applicant |
| EP0228242A2 | Cites | European Patent Office (EPO) | Applicant |
| EP0388658A1 | Cites | European Patent Office (EPO) | Applicant |
| EP0451384A1 | Cites | European Patent Office (EPO) | Applicant |
| EP0632382A1 | Cites | European Patent Office (EPO) | Applicant |
| EP0747841A1 | Cites | European Patent Office (EPO) | Applicant |
| CN101980174A | Cites | China | Applicant |
| US10219039B2 | Cites | United States of America | Applicant |
| US2001049620A1 | Cites | United States of America | Applicant |
| US2002040394A1 | Cites | United States of America | Applicant |
| US2002040395A1 | Cites | United States of America | Applicant |
| US2002049762A1 | Cites | United States of America | Applicant |
| US2002056087A1 | Cites | United States of America | Applicant |
| US2002072966A1 | Cites | United States of America | Applicant |
| US2002077787A1 | Cites | United States of America | Applicant |
| US2002078191A1 | Cites | United States of America | Applicant |
| US2002099812A1 | Cites | United States of America | Applicant |
| US2002099818A1 | Cites | United States of America | Applicant |
| US2002099819A1 | Cites | United States of America | Applicant |
| US2002103664A1 | Cites | United States of America | Applicant |
| US2002112048A1 | Cites | United States of America | Applicant |
| US2002124074A1 | Cites | United States of America | Applicant |
| US2002150054A1 | Cites | United States of America | Applicant |
| US2002161673A1 | Cites | United States of America | Applicant |
| US2003018969A1 | Cites | United States of America | Applicant |
| US2003046303A1 | Cites | United States of America | Applicant |
| US2003046385A1 | Cites | United States of America | Applicant |
| US2003062223A1 | Cites | United States of America | Applicant |
| US2003067554A1 | Cites | United States of America | Search report |
| US2003110485A1 | Cites | United States of America | Applicant |
| US2003144868A1 | Cites | United States of America | Applicant |
| US2003145319A1 | Cites | United States of America | Applicant |
| US2003149975A1 | Cites | United States of America | Applicant |
| US2003163563A1 | Cites | United States of America | Applicant |
| US2003182387A1 | Cites | United States of America | Applicant |
| US2003187677A1 | Cites | United States of America | Applicant |
| US2003208578A1 | Cites | United States of America | Applicant |
| US2003231203A1 | Cites | United States of America | Applicant |
| US2003233197A1 | Cites | United States of America | Search report |
| US2004003391A1 | Cites | United States of America | Applicant |
| US2004024717A1 | Cites | United States of America | Applicant |
| US2004221033A1 | Cites | United States of America | Applicant |
| JP2004357311A | Cites | Japan | Applicant |
| US2005114511A1 | Cites | United States of America | Applicant |
| US2005149964A1 | Cites | United States of America | Applicant |
| US2006074769A1 | Cites | United States of America | Applicant |
| US2006075421A1 | Cites | United States of America | Applicant |
| WO2006096177A1 | Cites | World Intellectual Property Organization (WIPO) | Applicant |
| US2006136965A1 | Cites | United States of America | Applicant |
| US2007011039A1 | Cites | United States of America | Applicant |
| US2007022032A1 | Cites | United States of America | Applicant |
| WO2007022250A2 | Cites | World Intellectual Property Organization (WIPO) | Applicant |
| US2007136753A1 | Cites | United States of America | Applicant |
| US2007143778A1 | Cites | United States of America | Applicant |
| US2007174295A1 | Cites | United States of America | Applicant |
| US2007276940A1 | Cites | United States of America | Applicant |
| US2007294729A1 | Cites | United States of America | Applicant |
| US2007294740A1 | Cites | United States of America | Applicant |
| JP2008011365A | Cites | Japan | Applicant |
| US2008077951A1 | Cites | United States of America | Applicant |
| US2008120650A1 | Cites | United States of America | Applicant |
| US2008140479A1 | Cites | United States of America | Applicant |
| US2008256235A1 | Cites | United States of America | Applicant |
| US2008263200A1 | Cites | United States of America | Applicant |
| US2008300965A1 | Cites | United States of America | Applicant |
| US2009055854A1 | Cites | United States of America | Applicant |
| US2009112703A1 | Cites | United States of America | Applicant |
| US2009147786A1 | Cites | United States of America | Applicant |
| US2010114527A1 | Cites | United States of America | Search report |
| US2010146533A1 | Cites | United States of America | Applicant |
| US2010205628A1 | Cites | United States of America | Applicant |
| US2010228855A1 | Cites | United States of America | Applicant |
| US2011004682A1 | Cites | United States of America | Applicant |
| US2011029636A1 | Cites | United States of America | Applicant |
| US2011208860A1 | Cites | United States of America | Applicant |
| US2011258049A1 | Cites | United States of America | Applicant |
| US2012124620A1 | Cites | United States of America | Applicant |
| US2012254911A1 | Cites | United States of America | Applicant |
| US2012260278A1 | Cites | United States of America | Applicant |
| KR20130008978A | Cites | Republic of Korea | Applicant |
| US2013160042A1 | Cites | United States of America | Applicant |
| US2013163658A1 | Cites | United States of America | Applicant |
| US2013179914A1 | Cites | United States of America | Applicant |
| US2013227595A1 | Cites | United States of America | Applicant |
| US2014150003A1 | Cites | United States of America | Applicant |
| US2014379421A1 | Cites | United States of America | Applicant |
| US2014380348A1 | Cites | United States of America | Applicant |
| US2014380350A1 | Cites | United States of America | Search report |
| US2015067075A1 | Cites | United States of America | Applicant |
| US2015189378A1 | Cites | United States of America | Applicant |
| US2015229979A1 | Cites | United States of America | Applicant |
| US2016037201A1 | Cites | United States of America | Search report |
| US2016092267A1 | Cites | United States of America | Search report |
| US2016165277A1 | Cites | United States of America | Search report |
| US2016165287A1 | Cites | United States of America | Applicant |
| US2016269783A1 | Cites | United States of America | Applicant |
7 members in 1 office; this record represents the family
Priority claims2
| Document | Office | Kind | Date |
|---|---|---|---|
| 201615385508 | United States of America | A | |
| US201615385508 | – | – | – |
Members7
| Document | Office | Kind | |
|---|---|---|---|
| US2018176622A1 | United States of America | A1 | |
| US10791355B2This record | United States of America | B2 | |
| US2020413116A1 | United States of America | A1 | |
| US11778255B2 | United States of America | B2 | |
| US2023412862A1 | United States of America | A1 | |
| US12356032B2 | United States of America | B2 | |
| US2025294202A1 | United States of America | A1 |
106 transactions on the USPTO file
Allowed after 3 non-final rejections, 2 final rejections and 1 RCE.
- Non-final rejections
- 3
- Final rejections
- 2
- RCEs
- 1
- Appeals
- 0
Over time
Point at a mark for the transactionTransactions
| Event | Code | |
|---|---|---|
| Payment of Maintenance Fee, 4th Year, Large EntityM1551 | M1551 | |
| Post Issue Communication - Certificate of CorrectionN423 | N423 | |
| Recordation of Patent Grant MailedPGM/ | PGM/ | |
| Patent Issue Date Used in PTA CalculationAllowedPTAC | PTAC | |
| Email NotificationEML_NTR | EML_NTR | |
| Issue Notification MailedAllowedWPIR | WPIR | |
| Dispatch to FDCD1935 | D1935 | |
| Application Is Considered Ready for IssuePILS | PILS | |
| Issue Fee Payment VerifiedN084 | N084 | |
| Issue Fee Payment ReceivedIFEE | IFEE | |
| Email NotificationEML_NTR | EML_NTR | |
| Printer Rush- No mailingTCPB | TCPB | |
| Mail Miscellaneous Communication to ApplicantMM327 | MM327 | |
| Miscellaneous Communication to Applicant - No Action CountM327 | M327 | |
| Information Disclosure Statement consideredIDSC | IDSC | |
| Pubs Case Remand to TCPUBTC | PUBTC | |
| Information Disclosure Statement (IDS) FiledM844 | M844 | |
| Information Disclosure Statement (IDS) FiledWIDS | WIDS | |
| Response to Reasons for AllowanceREAS | REAS | |
| Response to Reasons for AllowanceREAS | REAS | |
| Electronic ReviewELC_RVW | ELC_RVW | |
| Email NotificationEML_NTF | EML_NTF | |
| Mail Notice of AllowanceAllowedMN/=. | MN/=. | |
| Notice of Allowance Data Verification CompletedAllowedN/=. | N/=. | |
| Reasons for AllowanceEX.R | EX.R | |
| Interview Summary - Applicant Initiated - TelephonicEXAT | EXAT | |
| Information Disclosure Statement consideredIDSC | IDSC | |
| Date Forwarded to ExaminerFWDX | FWDX | |
| Response after Non-Final ActionA... | A... | |
| Request for Extension of Time - GrantedXT/G | XT/G | |
| Electronic ReviewELC_RVW | ELC_RVW | |
| Email NotificationEML_NTF | EML_NTF | |
| Mail Non-Final RejectionNon-final rejectionMCTNF | MCTNF | |
| Non-Final RejectionNon-final rejectionCTNF | CTNF | |
| Information Disclosure Statement consideredIDSC | IDSC | |
| Date Forwarded to ExaminerFWDX | FWDX | |
| Disposal for a RCE / CPA / R129AbandonedABN9 | ABN9 | |
| Electronic Information Disclosure StatementEIDS. | EIDS. | |
| Request for Continued Examination (RCE)RCEX | RCEX | |
| Information Disclosure Statement (IDS) FiledWIDS | WIDS | |
| Workflow - Request for RCE - BeginBRCE | BRCE | |
| Email NotificationEML_NTR | EML_NTR | |
| Mail Applicant Initiated Interview SummaryMEXIA | MEXIA | |
| Electronic ReviewELC_RVW | ELC_RVW | |
| Email NotificationEML_NTF | EML_NTF | |
| Mail Final Rejection (PTOL - 326)Final rejectionMCTFR | MCTFR | |
| Interview Summary - Applicant Initiated - TelephonicEXAT | EXAT | |
| Interview Summary- Applicant InitiatedEXIA | EXIA | |
| Final RejectionFinal rejectionCTFR | CTFR | |
| Information Disclosure Statement consideredIDSC | IDSC | |
| Date Forwarded to ExaminerFWDX | FWDX | |
| Response after Non-Final ActionA... | A... | |
| Information Disclosure Statement (IDS) FiledM844 | M844 | |
| Information Disclosure Statement (IDS) FiledWIDS | WIDS | |
| Electronic ReviewELC_RVW | ELC_RVW | |
| Email NotificationEML_NTF | EML_NTF | |
| Mail Non-Final RejectionNon-final rejectionMCTNF | MCTNF | |
| Non-Final RejectionNon-final rejectionCTNF | CTNF | |
| Date Forwarded to ExaminerFWDX | FWDX | |
| Response after Final ActionA.NE | A.NE | |
| Email NotificationEML_NTR | EML_NTR | |
| Application ready for PDX access by participating foreign officesCCRDY | CCRDY | |
| PG-Pub Issue NotificationPG-ISSUE | PG-ISSUE | |
| Electronic ReviewELC_RVW | ELC_RVW | |
| Email NotificationEML_NTF | EML_NTF | |
| Mail Final Rejection (PTOL - 326)Final rejectionMCTFR | MCTFR | |
| Final RejectionFinal rejectionCTFR | CTFR | |
| Information Disclosure Statement consideredIDSC | IDSC | |
| Information Disclosure Statement consideredIDSC | IDSC | |
| Information Disclosure Statement consideredIDSC | IDSC | |
| Electronic Information Disclosure StatementEIDS. | EIDS. | |
| Information Disclosure Statement (IDS) FiledWIDS | WIDS | |
| Date Forwarded to ExaminerFWDX | FWDX | |
| Response after Non-Final ActionA... | A... | |
| Information Disclosure Statement (IDS) FiledM844 | M844 | |
| Information Disclosure Statement (IDS) FiledWIDS | WIDS | |
| Information Disclosure Statement (IDS) FiledM844 | M844 | |
| Electronic Information Disclosure StatementEIDS. | EIDS. | |
| Information Disclosure Statement (IDS) FiledWIDS | WIDS | |
| Electronic ReviewELC_RVW | ELC_RVW | |
| Email NotificationEML_NTF | EML_NTF | |
| Mail Non-Final RejectionNon-final rejectionMCTNF | MCTNF | |
| Information Disclosure Statement (IDS) FiledWIDS | WIDS | |
| Information Disclosure Statement (IDS) FiledM844 | M844 | |
| Non-Final RejectionNon-final rejectionCTNF | CTNF | |
| Information Disclosure Statement consideredIDSC | IDSC | |
| Case Docketed to Examiner in GAUDOCK | DOCK | |
| Information Disclosure Statement (IDS) FiledM844 | M844 | |
| Information Disclosure Statement (IDS) FiledWIDS | WIDS | |
| Oath or Declaration Filed (Including Supplemental)C602 | C602 | |
| Miscellaneous Incoming LetterLET. | LET. | |
| Case Docketed to Examiner in GAUDOCK | DOCK | |
| Application Dispatched from OIPEOIPE | OIPE | |
| Email NotificationEML_NTR | EML_NTR | |
| Application Is Now CompleteCOMP | COMP | |
| Filing ReceiptFLRCPT.O | FLRCPT.O | |
| Application Is Now CompleteCOMP | COMP | |
| Sent to Classification ContractorPGPC | PGPC | |
| FITF set to YES - revise initial settingFTFS | FTFS | |
| Cleared by OIPE CSRL194 | L194 |
28 legal events, as the office reported them to INPADOC
Over the term
Point at a mark for the eventEvents
| Event | Code | |
|---|---|---|
| Maintenance fee paymentMAFP | MAFP | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| AssignmentAS | AS | |
| Certificate of correctionCC | CC | |
| AssignmentAS | AS | |
| Information on status: patent grantGrantedPATENTED CASESTCF | STCF | |
| Information on status: patent application and granting procedure in generalPUBLICATIONS -- ISSUE FEE PAYMENT VERIFIEDSTPP | STPP | |
| AssignmentAS | AS | |
| Information on status: patent application and granting procedure in generalAWAITING TC RESP., ISSUE FEE NOT PAIDSTPP | STPP | |
| Information on status: patent application and granting procedure in generalNOTICE OF ALLOWANCE MAILED -- APPLICATION RECEIVED IN OFFICE OF PUBLICATIONSSTPP | STPP | |
| Information on status: patent application and granting procedure in generalRESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINERSTPP | STPP | |
| Information on status: patent application and granting procedure in generalNON FINAL ACTION MAILEDSTPP | STPP | |
| Information on status: patent application and granting procedure in generalDOCKETED NEW CASE - READY FOR EXAMINATIONSTPP | STPP | |
| Information on status: patent application and granting procedure in generalFINAL REJECTION MAILEDSTPP | STPP | |
| AssignmentAS | AS |
Numbers
- Publication
- 10791355
- Publication, DOCDB
- 10791355
- Publication, EPODOC
- US10791355
- Application
- 15385508
- Application, DOCDB
- 201615385508
- Application, EPODOC
- US201615385508
Titles
- English
- Methods and apparatus to determine probabilistic media viewing metrics
Patent term adjustment
- A delay
- +32 daysthe office missed an examination deadline
- Applicant delay
- −137 days
- Net adjustment
- 0 days
Classification
- CPC, 3
- H04N21/25891
- H04N21/252
- H04N21/25883
- IPC, 5
- G06K9 00
- G11B27 10
- H04N21 258
- H04N21 25
- G06F19 20
- USPC, 1
- 382103000